57add9d754d5409aba8d9decfa7796b4

Priced Out

Fri, 18 Jul 2025 13:43:20 GMT

Affordability of house prices and rents worsening

Housing affordability in the UK has become increasingly challenging over the past two decades, affecting millions of households across the country. Rising house prices, escalating rents and relatively stagnant income growth have combined to create significant barriers to homeownership and financial stability.

Buying and renting has become less affordable

Since the late 1990s, the average price of a home in England and Wales has quadrupled, increasing from around £70,000 in 1998 to nearly £290,000 by 2024. In contrast, average incomes have grown far more modestly, resulting in a substantial widening of the affordability gap. By 2024, the median house-price-to-income ratio in England had risen to approximately 7.7, compared to around 4.5 two decades ago. However, these national averages obscure significant regional variations. In London, affordability is especially severe, with house prices frequently exceeding twelve times median local incomes, whereas in northern regions like Yorkshire and the North East, the ratio typically remains around five.

In parallel with soaring property prices, the private rental sector has also seen substantial increases. Over the past ten years, average rents in the UK have risen by nearly 46%, from around £924 per month to approximately £1,344 by mid-2025. Renters, particularly those in lower-income brackets, now spend a disproportionate share of their earnings on housing costs—often over 60% of their monthly income. Again, London renters face the harshest conditions, spending an average of nearly half their income on rent.

Government policy

These affordability pressures have significantly reshaped the UK's housing tenure landscape. Homeownership rates have notably declined, particularly among younger adults. In 2003, approximately 71% of UK households owned their home; by 2018, this had fallen below 64%. For young adults aged 16–29, homeownership dropped even more dramatically, from 21% in 1995 to around 10% today, pushing many into expensive and insecure rental situations.

Government policy over recent decades has attempted to address these challenges, yet with mixed outcomes. Historic measures, such as the Right to Buy scheme, while successful in expanding homeownership initially, significantly reduced the availability of social housing and inadvertently contributed to the current shortage. Similarly, the Help to Buy scheme, introduced in 2013, aimed to assist first-time buyers through equity loans and mortgage guarantees but largely had the unintended consequence of inflating house prices further.

Need for supply side balance

Recent policies have shifted towards addressing the immediate pressures facing renters and prospective buyers. Measures such as tenant protections, including the abolition of no-fault evictions and the introduction of rent controls, aim to increase security within the private rental sector. The government’s new permanent mortgage guarantee scheme and relaxed lending guidelines from institutions like Nationwide also attempt to ease entry barriers for first-time buyers. Additionally, regional interventions, such as council-tax premiums on second homes in parts of Wales, reflect attempts to manage local affordability crises.

In relation to the supply side, the government is committed to building 1.5m new homes by 2029. In addition, £39 billion has been allocated for the Social and Affordable Homes programme that could see 200,000 new social and affordable homes built. Some commentators, while welcoming the investment and targets, question whether this will be sufficient to make an impact on the scale of the affordability crisis especially as the Office for Budget Responsibility recently warned that housebuilding will fall short of target by about half a million new homes.

To meaningfully improve housing affordability, the UK government will need prioritise boosting housing supply, particularly affordable and social housing, implement effective planning reforms and strategically address regional disparities. Without comprehensive action tackling both supply and systemic inequalities, housing affordability will remain a severe challenge, perpetuating financial insecurity and intergenerational inequality for many years to come.

Why is Storytelling So Difficult?

Mon, 23 Jun 2025 11:36:04 GMT

And what can researchers do to get better at it?

For years now, the market research industry has championed storytelling as an essential skill, advocating a shift from purely reporting data to creating compelling narratives.

Despite persistent advocacy from senior figures, effective storytelling seems to be problematic for the industry with yet another prominent market research conference in June 2025 calling for more storytelling. Understanding why storytelling in market research is challenging and identifying the key elements of crafting impactful narratives from insight data, are crucial for influencing decision-making more effectively.

Why is it such a problem?

One reason why storytelling is difficult in market research lies in the complexity and detail within data. Researchers, trained in data accuracy, often find it challenging to distil extensive and nuanced information into concise, engaging narratives. This fear of oversimplifying critical details frequently leads to overly complex presentations that obscure the central message.

Another challenge arises from the mismatch between typical researcher skillsets and the skills required for storytelling. Researchers often excel analytically but can lack the narrative and creative skills necessary for engaging storytelling. Effective storytelling requires creativity, empathy and a deep understanding of audience psychology. These skills are not traditionally part of the training programme of a typical market researcher.

Organisational barriers further complicate effective storytelling. Many corporate cultures place greater value on objectivity and precision, inadvertently undermining narrative methods by viewing them as less rigorous. As a result, traditional data reporting methods continue to dominate, stifling innovation in how insights are communicated.

Understanding the audience

A fundamental issue underpinning these challenges is a lack of clarity about audience needs. Effective storytelling demands a clear understanding of the audience's priorities, motivations and challenges. Market researchers often struggle to sufficiently empathise with or comprehend the audience’s perspective because they do not have a clear understanding of their clients' business issues. This makes it very difficult to put together an insightful and engaging presentation that addresses the business issues. Being 'business savvy' should be a core part of a researcher's training.

What makes a good story?

A good story in the context of market research insight should address several crucial elements. Firstly, it should have a clear beginning, middle and end structure: beginning with context or challenge, moving onto data-driven insights and concluding with actionable recommendations. Nancy Duarte, in her influential work 'Resonate', underscores the necessity of structured narrative arc for communicating insights effectively.

In addition, emotional resonance is essential for impactful storytelling. Pure data rarely changes behaviour or attitudes; good storytelling engages emotions, making insights relatable and memorable. According to Chip and Dan Heath’s book "Made to Stick," stories that resonate are simple, unexpected, concrete, credible, emotionally engaging and deliver clear messages.

Contextual relevance is another critical aspect of effective storytelling. The insights presented should directly relate to the business issues (strategic and operational) faced by the audience, making the data meaningful and actionable within their specific context.

Really importantly, storytelling should explicitly encourage action. Effective narratives do not just highlight insights; they clearly indicate next steps and practical recommendations. Demonstrating how insights can tangibly influence decisions enhances their credibility and impact.

And finally, data visualisation

Finally, visualisation and creativity significantly enhance storytelling effectiveness. Thoughtful visualisations, such as infographics, interactive charts and dashboards, can intuitively communicate complex data, making it more accessible and engaging. Edward Tufte's work on data visualisation illustrates how well-designed visuals enhance comprehension and retention of information.

To achieve better storytelling, organisations should actively invest in developing researchers' narrative skills, encourage creative experimentation and cultivate a culture that equally values analytical rigour and engaging communication.

Only then will we put an end to the perennial call for more emphasis on storytelling.

Research Degrees

Sat, 21 Jun 2025 12:27:54 GMT

Do you need a degree to be a an insight professional?

The debate over whether a degree is essential to becoming a successful market researcher ongoing. Reflecting on my 40 years of experience both as a senior manager within research agencies and more recently as a client commissioning research,I believe that the core skills necessary for insightful market research do not inherently require a formal degree. However, degrees do often increase the likelihood of finding individuals who possess these valuable capabilities.

At its heart, market research demands skills that extend well beyond basic literacy and numeracy. Crucially, it requires deep curiosity, intellectual rigour, clear logic, adept storytelling, client negotiation skills and effective people management. These attributes enable researchers not just to collect data but to tease out the deeper insights and narratives that inform strategic decisions. Interestingly, many of these skills, such as curiosity and storytelling, are not easily measured or taught through formal education alone.

In addition, formal education is no guarantee of research competence. One of the most serious errors I encountered in my own career was made by someone with impeccable academic credentials - a PhD. It was me! This experience underscores the fact that degrees, even advanced ones, do not immunise researchers against critical mistakes.

As a senior manager in an agency, I observed that the more junior researchers often struggled with independent thought and initiative, potentially reflecting the limitations of contemporary degree programmes. Today's higher education system may inadvertently encourage a more passive approach to learning, driven by compliance and prescribed curriculum, rather than fostering independent, creative and critical thinking.

Looking forward, the evolution of generative AI in research will further reshape necessary skill sets. Researchers need to be able to ask the right questions, critically evaluate AI outputs and use technology ethically and effectively to deliver strategic insights. These skills again depend heavily on curiosity, discernment and intellectual flexibility - attributes not exclusively acquired through a formal degree.

In conclusion, while the skills critical for generating valuable insights do not strictly require a formal degree, having one certainly helps. Degrees often signal a level of rigour and discipline which correlate with strong analytical and intellectual capabilities.

However, the true measure of a market researcher lies in the broader set of personal and professional skills - many of which can indeed be nurtured outside of higher education.

The Survey Questioned

Thu, 19 Jun 2025 10:19:23 GMT

Has data analytics killed the survey?

Surveys have been a cornerstone of market research. Historically, they have provided meaningful insight for brands into crucial metrics such as awareness, consideration, intention and action.

While the survey methodology itself has evolved significantly, transitioning from face-to-face interviews and telephone surveys to sophisticated online platforms and advanced restech tools, the fundamental principle of directly questioning consumers about their familiarity with brands, their purchase intentions and post-purchase perceptions remains unchanged.

However, the application of data analytics and text analytics tools has led us to question whether the traditional survey is becoming redundant for brand metrics. These techniques enable publicly available datasets to be mined, unlocking brand insights previously accessible only through primary research. But can these methods fully replace the traditional survey?

Measuring Brand Awareness and Engagement

Today, social media platforms such as Twitter, Instagram and TikTok provide valuable opportunities for marketing departments to gauge brand awareness, campaign impact and depth of consumer engagement. On a basic level, comparative follower analysis provides quick insight into market positioning against competitors, while tracking follower growth over time can effectively measure the success of marketing campaigns.

Online review platforms such as Amazon, Trustpilot and TripAdvisor have become important sources of consumer feedback. Unlike surveys, which rely on participants’ recollection and self-reporting, consumer reviews capture nuanced insights from customers on product strengths, weaknesses and competitive positioning.

Advanced analytical techniques such as sentiment analysis and topic modelling offer marketers deeper and more nuanced insights from social media and review site data. Sentiment analysis applies machine learning algorithms to automatically interpret the tone and emotional context of posts and comments, providing marketers with a clear view of public perceptions and brand sentiment.

Topic modelling identifies key themes emerging from user-generated content, highlighting consumer priorities, preferences and emerging trends. These techniques enable marketers to understand not only who is engaging with their brand but also how they feel and what specific aspects resonate most strongly. Complementing this, tools like Google Trends track shifts in search volumes, revealing dynamic trends in consumer interest, engagement intensity and brand relevance.

The Continuing Value of Surveys

Despite these capabilities, news of the survey's death is premature. Surveys retain many strengths in relation to brand metrics that analytics and existing datasets cannot fully replicate. Specifically, surveys offer:

Detailed Brand Awareness Metrics: Surveys can precisely distinguish between spontaneous and prompted brand awareness, providing clarity on consumer recall and recognition.
In-Depth Attitudinal and Perception Data: Surveys enable marketers to explore what different consumer segments think about brands, products, or services and uncover specific usage contexts.
Comprehensive Engagement Metrics: Surveys can capture the frequency, motivation and touchpoints within the customer journey which can be used to provide predictive insights into intentions.

Additionally, surveys facilitate advanced statistical analyses such as key driver analysis and conjoint (trade-off) analysis, which identify consumer preferences on product features, pricing structures and promotional strategies which are critical insights for optimising product-market fit.

Integrated Insights for Complete Business Intelligence

We should not view techniques such as text analytics as a rival to the survey but rather as valuable complements to traditional survey methodologies. While analytics offer rapid, cost-effective and genuine consumer insight at scale, surveys provide essential detail, brand-specific depth and causal linkages missing from purely observational data.

The most effective strategy for a brand is to utilise an integrated research strategy, combining the immediacy and authenticity of data analytics with the robust depth and structured insights of surveys. By applying both approaches in tandem, brands can achieve comprehensive and actionable business intelligence, better positioning their brands in an increasingly competitive marketplace.

Fascinating Fakery

Mon, 02 Jun 2025 09:48:51 GMT

Is synthetic data the panacea for an industry struggling with data quality issues?

Market research is caught in a relentless race towards speed, affordability and scalability, driven by the increasingly urgent demands of modern business. However, this pursuit comes at a price: data quality. The industry today is littered with poorly engaged respondents, bots, rushed completions and outright fraudulent entries. Now, generative AI has entered the room, capable of creating convincing "synthetic respondents."

This begs a critical question: should market researchers embrace AI-generated responses or is this a Pandora's box best left unopened?

The data quality struggle

Traditional surveys have long struggled with data quality. Speeders who race through surveys without thoughtful consideration, straight-liners who select the same answer repetitively and random clickers who choose options without engaging, are not just annoyances, they actively undermine insight integrity. Many panellists are driven by incentives, turning data collection into a transactional, rather than reflective, exercise. Add to this the widespread issue of survey fatigue, exacerbated by lengthy, tedious or poorly designed questionnaires and the conditions are ripe for low-quality responses. These concerns become especially evident in open-ended questions, where copy-pasted answers and shallow insights often dominate. Before even considering AI-generated text, much of the qualitative input we currently receive is already alarmingly thin.

Enter generative AI

Enter generative AI, presenting itself as both intriguing and potentially problematic. Synthetic respondents are essentially AI-generated outputs that simulate human responses. These virtual participants can convincingly articulate answers to survey questions, extracted from extensive language models trained on vast datasets of human-produced text. Academia and commercial businesses have already begun experimenting with this technology, attracted by its substantial promise: faster turnaround times, dramatically reduced costs and the ability to simulate niche or difficult-to-reach populations effortlessly. In addition, synthetic respondents offer the unique ability to rigorously scenario-test surveys, ensuring robustness before deployment to real-world audiences.

Ethical minefield and bias

However, synthetic data represents an ethical and methodological minefield. Central among these issues is validity: can an AI truly replicate nuanced human behaviour, complete with emotional responses, subtle biases and unpredictable cognitive patterns? AI-generated outputs might sound plausible but they lack genuine lived experiences and authentic emotional engagement. Furthermore, transparency becomes paramount. Shouldn't clients and stakeholders be explicitly informed when synthetic data has influenced - or replaced - real participant responses? Without transparency, trust and credibility in market research could be severely compromised.

Bias amplification poses another significant risk. AI models inherit and reflect the biases present within their training datasets. If unchecked, synthetic respondents might unwittingly perpetuate or even exacerbate existing stereotypes, inadvertently shaping business strategies around fundamentally flawed insights. Equally troubling is the risk of overconfidence. Beautifully constructed but ultimately ungrounded responses can create a dangerous illusion of insight, potentially leading researchers down misguided strategic paths.

Legitimate uses

There are, however, legitimate and constructive uses for synthetic data, provided clear boundaries are maintained. Pre-testing survey logic or refining wording is an ideal candidate, offering a practical, low-risk entry point into synthetic applications. AI-generated responses can also be beneficial for creating "expected" outputs that train or validate analytical models. Additionally, synthetic respondents can spur creativity and idea generation in the exploratory phases of qualitative research.

But there are clear ethical lines that must not be crossed. Replacing genuine respondents entirely in quantitative studies is deeply problematic, undermining the very essence of market research - authentic human understanding. Equally concerning is mixing real and synthetic data without disclosure, a practice that risks deceit and erodes stakeholder trust. Using synthetic data to artificially "validate" findings also threatens methodological integrity, distorting reality and potentially leading to misguided business decisions.

The implications for the market research industry

The rise of synthetic data carries profound implications for market research professionals. AI literacy is increasing rapidly. The industry needs to take stock and think about how to critically assess AI-generated outputs, distinguishing between genuine human insight and artificially intelligent mimicry. Data quality needs reframing: beyond merely eliminating poor respondents, researchers must now rigorously question the origins and authenticity of every data point. This evolution necessitates updated industry standards and ethical guidelines, explicitly addressing the legitimate use and disclosure of synthetic data.

While caution is essential, outright dismissal of synthetic respondents should be avoided - for the moment at least. Generative AI undoubtedly holds potential for innovation and efficiency in market research. The challenge lies in harnessing this potential responsibly. The industry should proceed with a balanced mix of curiosity and vigilance, prioritising transparency, rigorous methodology and the preservation of trust above all else.

Is Big Tech Eating Our Lunch?

Mon, 26 May 2025 19:30:13 GMT

Do tech people know more about market research than social scientists?

Recently, I came across a sentiment that resonated deeply: "We live in an age where tech people think they know more about market research than a sociologist."

Undeniably, technology has revolutionised market research methodologies, providing tools of unprecedented power and precision. Data science, artificial intelligence and machine learning promise almost instant insights, distilling vast data sets into concise, actionable strategies. As technology companies move into areas traditionally the province of market research, it might appear at face value that the researcher's traditional expertise - the subtle, nuanced understanding of human behaviour - is becoming sidelined.

First principles

I'm not so sure. The foundational principle of market research has not fundamentally changed. At its heart, effective research is not solely about data; it is about insight and understanding the nuances, contradictions and subtleties of human behaviour. Here, human understanding remains irreplaceable. While algorithms can identify patterns within data, researchers interprets the significance of those patterns within a cultural and emotional context.

This brings me to a vital realisation that while technology undeniably enriches our toolkit, sociological understanding ensures we remain connected to the complex, emotional and often unpredictable human factors that drive true consumer insight.

The human factor

Algorithms, as advanced as they have become, do not yet comprehend irony, hidden social pressures or cultural subtleties. Human life rarely fits neatly into binary code or statistical categories.

But ignoring technological advances would be equally misguided. The data-driven precision that technology offers is transformative. It complements rather than replaces human insight, allowing us to test hypotheses, validate intuitions and deliver sharper, more meaningful analysis.

I believe that the integration of human insight with technological capabilities is a powerful combination and one that underpins our business philosophy. Technology enriches our methodological toolkit but sociological understanding provides essential context and meaning

Does A Personal Brand Matter?

Mon, 12 May 2025 11:00:12 GMT

Why branding is no substitute for authenticity

I was once told that I needed to 'work on my personal brand.'

I ignored the advice because the concept felt uncomfortable, superficial even. Having spent four decades in the insights industry, climbing the ladder into senior leadership - although never quite ascending to the very top - I wonder whether my lack of a personal brand explains why I've fallen short of those top-tier roles.

Reflecting on my career, I see patterns that perhaps limited my ascent to the pinnacle. My leadership style was never loud or flashy. Instead, I quietly encouraged and mentored those around me. I derived satisfaction from seeing colleagues do well and didn't see the need to blow my own trumpet. Was this a mistake? Should I have embraced a more assertive or self-promoting strategy?

Also, the thought of spending evenings schmoozing at industry dinners or networking drinks filled me with dread and I'd much rather return home to my family. In those quiet choices, perhaps I unwittingly defined my personal brand without realising it: competent, supportive, reliable but doesn't have what it takes to get to the top.

I do sometimes ask myself if actively managing my personal brand would have taken my career further? Professional life nowadays is seemingly obsessed with personal branding. LinkedIn bios promise all sorts of dynamic impact with people shouting from the virtual rooftops about accomplishments both genuine and exaggerated. We are encouraged to project a meticulously crafted image, strategically aligned across our digital and real-world presences. But does it really matter?

There's something undeniably uncomfortable about reducing expertise and character to a glossy personal branding exercise. After all, authenticity should surely speak louder than carefully scripted LinkedIn posts? But if authenticity alone were enough, would I even be asking myself these questions now?

Perhaps personal branding does matter. Perhaps it's not vanity but a means of taking control of your own professional narrative rather than leaving it to chance or assumption. It might have stopped me from being pigeon-holed as quiet or peripheral, qualities that could have obscured my genuine influence and expertise. It might have helped me build a clearer pathway to that elusive top job.

But equally, I question whether relentless personal branding risks overshadowing the deeper values that shape a genuinely meaningful career. Integrity, professionalism and a genuine care for others rarely make for captivating social media posts or headline-worthy LinkedIn profiles. Yet they define the legacy of a leader more profoundly than any flashy brand narrative ever could.

So perhaps, after all these years, the real question isn't whether a personal brand is important but what is it exactly that personal brand represents? For me, it's been quiet strength, consistency, dedication and a belief that one's life is not solely defined by professional triumphs but by balance and personal values as well.

Could I have achieved more if I'd networked harder or shouted louder? Probably. But I'm proud that I led with integrity and care and not through self-aggrandisement.

In the end, your personal brand probably does matter. But perhaps it should reflect who you genuinely are and not what others expect you to project.

Maybe authenticity, ultimately, is the most powerful branding strategy of all.

Survey Respondents Lack Identity

Thu, 08 May 2025 10:09:09 GMT

Shorter surveys won't solve fake answers

The recent exposure of fraudulent survey practice at an agency in the USA has once again put the spotlight on the reliability of survey data. Some industry commentators take the view that agencies are actually very good at anti fraud actions and that the real issue affecting data quality is long and boring surveys.

It is true that excessively long and boring surveys significantly contribute to respondent fatigue and poor-quality data. However, the industry's current focus on survey length, question engagement and commoditisation of sample overlooks a more fundamental issue: respondent authenticity. While improving survey design is undoubtedly crucial, placing primary emphasis on outputs without first addressing the quality of inputs, specifically, respondent credibility, risks missing the core challenge.

Surveys that extend beyond reasonable lengths, filled with tedious questions presented in repetitive grids, undeniably encourage disengagement. Respondents confronted with such tasks naturally tend towards shortcuts, guesswork or complete abandonment. It is important that the industry moves away from lengthy, uninspiring questionnaires towards shorter, more engaging formats. However, to assume that halving survey duration and offering higher incentives will automatically resolve all data quality issues is overly simplistic.

The root cause of poor survey data often lies deeper, within the respondents themselves. A significant portion of respondents engage in practices that compromise data integrity either by gaming the system to maximise incentives, conforming their answers to perceived expectations of researchers or projecting an idealised persona rather than their true selves. These practices are not necessarily driven by survey length alone but reflect broader behavioural patterns and motivations that persist irrespective of questionnaire design.

Online survey panels often inadvertently incentivise dishonest or inaccurate participation. Low incentives might discourage authentic respondents, while simultaneously attracting those with minimal regard for data authenticity. Yet increasing payments alone will not necessarily filter out individuals intent on misrepresentation; indeed, it might attract more sophisticated forms of gaming. This suggests a critical oversight: sample quality, rather than simply sample cost or survey length, demands greater scrutiny.

Efforts to combat this issue have typically focused on sample providers’ existing fraud detection mechanisms and engagement algorithms. While these are essential, there is a need for more nuanced strategies to validate respondent authenticity proactively. Advanced behavioural analytics, sophisticated identity verification techniques and enhanced profiling methods could help distinguish genuine respondents from those presenting contrived or fantastical personas.

Furthermore, the industry must confront its reluctance to recognise this uncomfortable truth: some respondents simply invent data. They may construct imaginary lifestyles, opinions or purchasing habits, driven by internal narratives or a desire to meet perceived researcher expectations. Merely shortening the length of the survey or raising incentives will not eliminate this phenomenon. It requires concerted efforts at an industry-wide level to rigorously vet respondent credibility and continuously refine methodologies that flag and exclude unreliable data sources.

In conclusion, while reducing survey length and rethinking incentive structures are important steps towards improving data quality, the industry must first prioritise the quality and authenticity of respondents themselves.

Without addressing this foundational issue, attempts to improve outputs - however well-intentioned - will inevitably fall short. Ultimately, the path to better quality data involves deeper investment in understanding and verifying the legitimacy of those providing it.

Only then can survey data consistently deliver accurate, reliable and actionable insights.

Beyond The Stars

Wed, 07 May 2025 16:43:54 GMT

Why it's important to analyse customer reviews

Ratings and review sites like Trustpilot and TripAdvisor play an important part in consumer purchase decisions. Consumers read reviews to work out if they are likely to get what they pay for and whether it meets their needs. Brands also rely on these scores in their marketing to show how well they are trusted. How many adverts do you see where a brand gives its Trustpilot score as a way of instilling confidence in its products?

However, the simplicity of star ratings can often obscure the full customer experience. By focusing solely on these numerical metrics, brands risk missing critical insights that lie within the accompanying written reviews.

The whole story?

A five-star rating, for example, might seem like an excellent endorsement of a product or service. But without exploring the reviewer’s comments, brands may overlook subtle issues or areas for improvement. A customer may award top marks while expressing concerns about an aspect of their experience that didn’t detract from their overall satisfaction but could become a problem if left unaddressed. For example, a five star TripAdvisor review might mention how much the reviewer loved their hotel stay, yet highlight that the check-in process felt unnecessarily slow. Without analysing the review, the hotel would remain unaware of an operational inefficiency that could negatively impact future customers less forgiving than the one who wrote the review.

Similarly, lower ratings often fail to tell the whole story. A one-star review might suggest serious dissatisfaction but the accompanying explanation could reveal it was the result of a one-off issue rather than a systemic problem. Without this context, brands might overreact or misdirect resources attempting to address perceived problems that don't actually exist. For example, a Trustpilot reviewer might give a meal kit company a one star review but the real cause of their dissatisfaction isn't the quality of the food but the fact that the delivery was and hour late.

Stars and their causes

Another significant limitation of numerical ratings is their failure to capture the causes of their overall score over time. Written reviews, on the other hand, can offer valuable insights into evolving trends, emerging pain points or changes in customer expectations. Tracking the language and themes within reviews allows brands to anticipate problems before they become widespread or identify opportunities to differentiate their offerings.

It's also important to acknowledge that ratings on sites like Trustpilot and Amazon tend to coalesce at the top and the bottom of the scale with fewer 2,3 and 4 star ratings. But these middle ratings can actually be more important than five star or one star reviews because they offer more nuanced insight into areas for improvement.

How to stay ahead

The true strength of written reviews lies in their ability to reveal not just what customers think but why they feel that way. This is incredibly powerful for brands looking to understand the underlying motivations, frustrations and desires that shape customer behaviour. Advanced text analytics tools make it possible to analyse these reviews at scale, identifying patterns and themes that would be impossible to discern manually. For marketing teams, this means being able to craft more resonant messaging and campaigns. For insight teams, it provides a deeper layer of understanding to inform strategic decisions.

Relying exclusively on star ratings risks reducing the complexity of customer experiences to an oversimplified number. The interplay between ratings and reviews is where the real value lies.

By moving beyond the stars and unpacking the meaning behind customer ratings, brands can stay ahead of the market, address potential issues before they become problematic and create experiences that encourage customer loyalty.

From Description to Prediction

Mon, 28 Apr 2025 10:42:38 GMT

Time to Ditch The Descriptive Dashboard

Historically, marketing and communications teams have depended on dashboards packed with data about past performance. While these descriptive insights clearly show what has already happened, they offer limited guidance for future strategies.

Today, brands are rapidly adopting predictive research tools, powered by advances in machine learning and automation, which enable them to anticipate and swiftly respond to consumer behaviour and market trends.

From Description to Prediction

This shift reflects the broader changes in consumer expectations in today's connected world. Consumers now expect highly personalised experiences and immediate interactions from brands. Traditional dashboards, though informative, simply can't meet these real-time expectations. As a result, businesses are turning to advanced predictive platforms that use machine learning algorithms to forecast future consumer actions, anticipate emerging trends and proactively manage risks such as customer churn.

Modern machine learning models can interpret vast amounts of data, revealing hidden patterns and continuously refining their predictions with new information. For example, predictive analytics can help brands forecast seasonal demand shifts, identify changing consumer preferences early or detect competitive threats as they emerge. This foresight enables brands to act swiftly and decisively, maintaining a crucial advantage over competitors.

Predictive analytics also transforms campaign planning and content personalisation. Rather than replicating past successes, predictive decision engines empower marketers to craft highly targeted messages tailored precisely to the right audience at exactly the right moment. This degree of personalisation ensures that campaigns remain relevant and impactful, significantly improving consumer engagement and overall campaign results.

Real-Time Personalisation

Predictive analytics is not limited to planning; it also enhances real-time personalisation. Platforms equipped with automated insights can analyse data continuously, adapting content and messaging dynamically in response to immediate consumer interactions. This real-time agility allows brands to quickly adjust their campaigns, maximising engagement, driving higher conversion rates and deepening customer loyalty.

To achieve these results, brands increasingly rely on tools that blend artificial intelligence, natural language processing and advanced analytics. These tools typically feature streamlined data integration from multiple sources, automatic identification of significant data patterns, continuous model training, and seamless, real-time updating. Importantly, these advanced analytics tools are designed for ease of use, enabling marketing and insights teams without deep technical backgrounds to harness predictive insights effectively.

The New Competitive Advantage

Ultimately, the move from descriptive dashboards to predictive decision engines represents more than just technological evolution; it signals a profound shift in how brands gain competitive advantage. By prioritising predictive insights, brands can proactively manage market dynamics, optimise campaign performance and deliver genuinely personalised experiences to consumers.

Predictive analytics, powered by machine learning and automation, is no longer simply beneficial, it is becoming essential for brands aiming to lead their markets.

A Question of Trust

Mon, 28 Apr 2025 10:15:20 GMT

Why ID verification is a must for online surveys

The golden age of the online survey is well and truly over. A mixture of fraudulent responses and the shedding of senior expertise within agencies, means that it's too much of a risk to make business decisions based solely on survey data.

Bot networks exploiting panel based survey research is a well-known channel of fraudulent data but real panellist giving made up answers is a serious problem that the industry is only just waking up to.

Is size everything?

How has the industry got into this situation? For far too long, research providers have been concerned more about the size of their panel rather than its quality. It's a simple numbers game - the more respondents you have, the more surveys you can field and the more data you have on people, including their likes, dislikes, lifestyles and attitudes. And this had driven acquisitions and panel building with the message that size really does matter when it comes to panel.

But actually size isn't everything. Quality is. Providers are very happy to tell you how many people are on their panel but less happy to tell you about how many they can actually verify as real people. If you can't prove that panellists are who they say they are, then how can you be sure that their answers are true?

Fraudulent panellists are not solely motivated by money - let's face it, nobody is going to get rich filling in online surveys. No, what really motivates them is creating a false narrative around their identity and lifestyle that is reflected in how they answer questionnaires. Sometimes their inconsistencies can catch them out but more often than not they go under the radar because agency quality control systems are geared to catching speedsters and straightlining rather than unconsidered and unreal responses.

And because these people sign up to multiple panels, the problem is amplified within the industry.

Dilution of expertise

At the same time many market research agencies have dispensed with senior staff in order to reduce costs. This has compounded the problem. Senior staff have the experience to spot anomalies in the data, to question whether the findings makes sense and to push back on things that don't look right. This expertise is being lost and consequently agencies are becoming too junior in composition to spot fraudulent data.

The need for verification

So what can be done about it? Obviously, you can set traps within surveys to expose inconsistencies but the most effective thing to do is to stop these people joining in the first place by making it much harder to join.

A system of verification such as showing government ID (passport, driving licence or voter ID), plus an in-home sign up interview (including the collection of key demographics) would simultaneously deter fraudsters and make panels more believable. We will end up with panels being a lot smaller and the sign up process being more time consuming but at least they would be believable.

Ultimately, clients have to be able to trust the data. At the moment making business decisions based on survey data is just too much of a risk.

Making Sense of Volume Data

Tue, 08 Apr 2025 10:06:07 GMT

Using text analytics to cut through the noise

Brands have access to an ever-growing volume of customer feedback. Reviews, comments, emails and survey responses are available in vast quantities, creating a dense landscape of information. While this constitutes a rich source of potential insights, the sheer volume poses significant challenges. The primary issue is not simply collecting feedback but understanding how to extract meaningful insights from what initially appears as overwhelming noise.

Using text analytics to cut through

Navigating high-volume feedback requires careful consideration and sophisticated methods. Human effort alone quickly reaches limitations, as manually processing thousands of reviews or comments is impractical, costly and subject to biases and errors. Traditional manual analysis risks overlooking subtleties or emerging trends, limiting an organisation's responsiveness to evolving customer needs and sentiments.

This is where text analytics comes into play, providing organisations with the capability to systematically transform vast amounts of qualitative data into actionable insights. At its core, text analytics employs natural language processing (NLP), which interprets, categorises and quantifies textual data at scale. By using text analytics we can identify recurring themes, sentiments and patterns within customer feedback, making sense of what initially seems incoherent and daunting.

Effective use of text analytics begins with recognising its capabilities and limitations. It does not replace human input but enhances it by revealing patterns quicker than can be achieved manually. For example, analytics can detect subtle shifts in customer sentiment before they become critical issues. By tracking sentiment trends, brands can proactively address customer dissatisfaction, avoiding broader reputational damage.

Furthermore, text analytics can identify specific features or service elements frequently mentioned in feedback, offering precise, detailed insights. Brands can clearly see which aspects of their offerings resonate positively with customers and which may require attention or improvement. This granularity allows organisations to make targeted, informed decisions about product development, customer service enhancements and overall strategy.

Humans in the loop

However, implementing text analytics is not without challenges. Context matters significantly and analytical models must account for nuances such as sarcasm, cultural expressions or industry-specific jargon. Ensuring models are continuously refined through iterative training and validation is vital in maintaining accuracy and relevance.

Additionally, organisations should not become overly reliant on automated systems. While text analytics can highlight trends and areas for action, human interpretation remains essential in determining the best responses and strategies. A thoughtful blend of automated analysis and human insight ensures that insights drawn from feedback remain meaningful and actionable.

In conclusion, managing high volumes of feedback through text analytics provides a powerful means for organisations to extract clarity and value from what initially seems like noise. Embracing these technologies thoughtfully and recognising their complementary role alongside human judgment can transform overwhelming quantities of feedback into clear pathways for improvement and growth.

AI Snake Oil

Mon, 10 Mar 2025 12:15:03 GMT

AI based consumer insight models more illusion than substance

Market research has always sought reliable ways to understand consumer behaviour. With the rise of artificial intelligence (AI), many agencies now promote AI-driven models as revolutionary solutions, promising deep consumer insights at unprecedented speed and scale. However, caution is required. Despite appealing marketing, many AI-based market research solutions today echo the practices of historical snake oil salesmen, offering more illusion than substance.

Synthetic solutions

The foundational challenge with AI-driven consumer insights lies in the quality and representativeness of the training data. AI models learn patterns from historical data sets, which themselves are often skewed, incomplete or biased. When consumer insights rely on such data, the resulting analysis inherits these limitations. Many datasets exclude certain demographics, neglect cultural nuances or reflect outdated behaviour patterns. Consequently, AI models frequently deliver insights that are accurate in theory but irrelevant or misleading in practice.

Another issue is the incorporation of synthetic data—artificially created data points intended to fill gaps or enhance datasets. While synthetic data can superficially improve model performance, it also introduces distortions. Models trained partly on synthetic data often produce outputs detached from real-world contexts. Rather than providing nuanced insights into genuine consumer behaviour, these models deliver broad generalisations or, worse, misleading predictions that businesses mistakenly trust.

Patterns not perspectives

AI-driven consumer insight often struggles to produce genuinely new or innovative perspectives. The nature of AI, being reliant on historical patterns, means these models largely recycle existing knowledge. Rather than uncovering groundbreaking consumer motivations or future trends, AI models typically reinforce established observations. While useful for affirming known insights, this rarely helps organisations anticipate truly novel consumer shifts or disruptions.

Perhaps the most fundamental limitation is the misguided belief that consumer behaviour can be fully quantified, modelled and predicted through algorithms alone. Human decisions and behaviours are influenced by complex emotions, spontaneous circumstances, cultural contexts and ever-changing social dynamics. Algorithms, however sophisticated, cannot fully replicate or predict this nuanced complexity. AI-driven insights can therefore offer partial views but should never be mistaken for complete or infallible consumer understanding.

Properly utilised AI can make a difference

This does not render AI irrelevant in market research. Properly utilised, AI can efficiently manage large datasets, identify broad trends and automate repetitive tasks, freeing researchers to explore deeper human contexts. However, agencies promising that their AI alone will deliver transformative consumer insights risk selling false assurance, reminiscent of historical snake oil salesmen.

Genuine understanding of consumer behaviour remains deeply human, requiring context, intuition and exploration beyond algorithmic reach.

Has Data Science Finished Off the Survey?

Mon, 03 Mar 2025 14:59:32 GMT

Hybrid data is the way forward

In a previous article, we argued that online surveys are compromised through poor quality . With advanced analytics able to give critical business insight from existing data, has data science finished off the online survey? We look at the issues.

The evolution of data collection

Data science has revolutionised the way we collect, analyse and interpret data. With advanced algorithms, machine learning and big data analytics, marketers now have unprecedented access to real-time insights from vast amounts of information. Social media interactions, website analytics and transactional data provide a continuous stream of behavioural data that can be mined for patterns and trends. This shift allows for a more dynamic and comprehensive understanding of consumer behavior, potentially rendering traditional surveys outdated.

The reduced importance of surveys in the age of data

Surveys have long been a cornerstone of market research, providing direct insights into consumer attitudes, preferences, and intentions. However, the limitations of surveys are becoming more pronounced in the era of data science. Surveys often suffer from biases, low response rates and inaccuracies due to self-reporting. Additionally, they typically provide a snapshot in time, which can quickly become outdated in a fast-paced market environment.

Brand metrics in a data-driven world

Key marketing metrics such as brand awareness, consideration, purchase, and loyalty have traditionally been gauged through surveys. Brand awareness surveys ask consumers if they recognise a brand, consideration surveys gauge if a brand is in the running for a purchase, purchase surveys identify actual buying behaviour and loyalty surveys assess repeat purchase intentions.

With data science, these metrics can now be measured more accurately and in real-time. For example, brand awareness can be inferred from social media mentions and search engine queries. Consideration can be tracked through website behaviour and product comparisons. Purchase data is readily available from transactional records and loyalty can be monitored through repeat purchase patterns and customer lifetime value analysis.

The persistent value of surveys

Despite the advances in data science, surveys are not entirely dead. They continue to offer unique advantages that automated data collection cannot fully replace. Surveys allow for direct questioning about specific topics that might not be evident from behavioural data alone. They can capture qualitative insights, such as emotional drivers, perceptions and motivations, which are crucial for understanding the "why" behind consumer actions.

Further, surveys provide a way to test hypotheses and validate findings from data analytics. For instance, if transactional data shows a decline in repeat purchases, a survey can help uncover the underlying reasons, such as dissatisfaction with a product feature or customer service issue.

Integrating surveys and data science

The optimal approach for modern marketing strategies is to integrate surveys with data science. By combining direct feedback from surveys with the rich, continuous data provided by analytics, marketers can gain a holistic view of consumer behavior. This integration enables a more nuanced understanding of brand metrics and enhances the accuracy of insights.

For example, a comprehensive brand health study might use survey data to assess overall brand perception while leveraging social media analytics to gauge real-time brand sentiment. Purchase intent data from surveys can be cross-referenced with actual sales data to identify discrepancies and opportunities for intervention.

Conclusion: A hybrid future

In the age of data science, declaring the survey dead would be premature. Instead, surveys should evolve and adapt to complement the wealth of information provided by modern data analytics. For senior marketing managers, the challenge lies in balancing these tools to derive the most actionable insights. By embracing a hybrid approach, combining the strengths of both surveys and data science, marketers can achieve a deeper, more comprehensive understanding of their customers, driving more effective strategies and better business outcomes.

Where Do We Go From Here? The Agency Dilemma

Mon, 24 Feb 2025 12:04:36 GMT

Rediscovering the human touch is vital for the market research industry

Every article or LinkedIn post you read about market research talks about how the old ways don't work anymore and the new ways (predominantly AI driven) are the way forward.

The integration of Artificial Intelligence (AI) has indeed revolutionised processes, making research swifter and more efficient. However, amidst this whirlwind of digital transformation, a fundamental question arises: Are we losing the essence of market research in our quest to be relentlessly modern?

The AI revolution in market research

The advent of AI has been nothing short of transformative for market research. Its capabilities to expedite research, automate report writing, undertake complex analyses and even generate primary data are undeniable. The allure of AI lies in its promise of unparalleled speed and efficiency. Market research agencies have eagerly embraced these tools, each proclaiming their unique edge through AI integration.

Yet, as more agencies jump onto the AI bandwagon, an ironic sameness has begun to permeate the industry. The promise of differentiation through technology paradoxically leads to a homogenisation of services. In their rush to showcase AI-driven uniqueness, agencies risk becoming mere echoes of each other, distinguishable more by the AI tools they employ than by the intrinsic value they provide.

The homogenisation dilemma

The core of the issue lies in differentiation or the lack of it. In a market where every player is leveraging similar AI tools, the unique selling propositions that traditionally set agencies apart become blurred. AI, for all its analytical might, cannot replicate the nuanced understanding, the human touch and the relational depth that have long been the hallmarks of successful market research.

This overreliance on technology leads to a commoditisation of research services. When every agency offers AI-driven insights, the distinction shifts from the quality of insight to the efficiency of process. This is a precarious path, as it undermines the value of deep, insightful research in favour of speed and cost-effectiveness.

Going backwards to move forwards

The future of market research in the UK, therefore, might lie in an unexpected direction – going backwards to move forwards. This is not to suggest a rejection of technology, but rather a recalibration of focus. The emphasis needs to shift back to the core elements that have always driven effective market research: people, knowledge and client relationships.

People, with their ability to understand, empathise and connect, remain at the heart of effective market research. Knowledge, gleaned not just from data but from experience, intuition and understanding of human behaviour, is invaluable. The depth and richness of insights derived from human analysts cannot be fully replicated by AI.

Great client relationships are built on trust, understanding and a shared journey towards solutions. These relationships go beyond mere transactions; they are partnerships fostered through personal interaction, tailored service and a deep understanding of client needs.

Conclusion

In conclusion, while AI is an invaluable tool for enhancing efficiency and augmenting capabilities in market research, it should not become the be-all and end-all. The UK market research industry must not lose sight of its foundational pillars – people, knowledge and relationships.

In an age where technology is ubiquitous, the true differentiator will be the human element. It is by going 'old school' in valuing these elements that market research can truly advance into a future that is not just technologically rich but also humanly profound.

Plugging The Online Survey Reliability Gap

Wed, 12 Feb 2025 16:57:22 GMT

How to make online surveys more reliable

In our last article, we discussed how online surveys suffer from a large quality deficit . In this follow up article, we talk about the things you can do to make your survey data more usable.

The online survey quality deficit

To recap, there are three main weaknesses with online surveys:

They are not representative.
Respondents don't always tell the truth.
People sign up to multiple panels amplifying the problem with fake and ill considered answers.

So does this mean you have to ditch that expensive survey you've just commissioned? No but you need to understand the limitations of the data before making business decisions.

It starts at the beginning

It may seem obvious but it needs saying - only choose a reputable online survey provider. And by this I mean one that has verified that their panel members are who they say they are, have robust policies for dealing with fake, inconsistent and ill considered answers and who can tell you what proportion of responses are deleted for failing quality checks.

So let's assume that you've got the survey data. What happens now?

Triangulation is the key

Online surveys will give you a sense of where things are at and are indicative of views rather than a precise measurement of them. For many organisations, that is sufficient for their needs but if business critical decisions are to be made off the back of the data then more work on validation is required.

Triangulation is key here. Triangulation is a method used to increase the credibility of research. In this context it means comparing the survey results against other data to see whether the narrative is the same. Comparing against known facts is the most obvious thing to do. This could include comparing sales and frequency of purchase data from the research against known data from your CRM.

Publicly available data can also be used to triangulate. For example, using Google Trends to measure movements in pre and post campaign searches against survey data on awareness and subsequent engagement. Do they go in the same direction or are there differences that need further investigation? At a more basic level it could involve comparing trends in social media reach and engagement against brand metrics from the survey. Topic analysis of user reviews can be used to validate issues emerging from the survey and to double check survey data on flows in and out of a brand.

Our article here gives further information on how existing data sources can be used to obtain data on purchase funnel and wider market metrics.

The world of market research has become ever more agile and the ability to get 1,000 completed interviews by lunchtime is great for clients needing quick data upon which to act. Speed, however, must not come at the expense of quality. Poor data is poor data no matter how speedy its assembly.

Clients need to keep two things in mind when using online survey research. First, be aware of the limitations of the data as it will not be that precise. Second, verify against other sources before making business critical decisions.

Who'll Win the Premier League 2025?

Thu, 06 Feb 2025 11:24:39 GMT

Liverpool has 89% probability of winning the Premier League

With the January transfer window now shut, we have run a predictive model to see who will win the Premier League title, which teams will qualify for Champions League football and which will be relegated.

Our model predicts that Liverpool will win the title from Arsenal. Liverpool are clear favourites with an 89% probability of winning. There is an 80% probability that Arsenal will finish second.

In relation to qualifying for the Champions League, Nottingham Forest has a 61% chance of finishing in the top four with Manchester City on 60%. If the current Premier League holders continue with their stuttering form then Chelsea (39% to finish in top four) and Newcastle (22%) could take their place.

At the bottom, Southampton are certain to be relegated and there is a very high probability that Ipswich and Leicester will join them. Wolves, with a 21% chance of finishing in the bottom three, are not out of the woods yet.

Premier League Final Table - Percentage Probability of Finishing In Each Position

Our predicted final Premier League standings is based upon probabilities. To generate the probabilities we use results from this season and previous ones to model the number of goals scored in a match as a Poisson distribution (the probability for a team to score one goal, two goals, three goals and so on in a match). We then identify parameters to represent a team’s attacking strength, defensive strength and home advantage. Applying this model to the remaining Premier League fixtures we can predict the final standings probabilities.

Striking The Right Chord

Wed, 05 Feb 2025 10:05:48 GMT

Applying sentiment analysis to understand how customers really feel

So far in our series on text analytics we have discussed the importance of looking beyond the number of stars you receive for your product or service by applying text analytics to analyse reviewers' comments. In this article, we take things a stage further by discussing the importance of sentiment analysis in understanding what customers really feel.

For brands navigating consumer review sites like Trustpilot, Amazon and TripAdvisor, the challenge isn’t just keeping up with what people are saying; it’s understanding how those words reflect the underlying sentiment. This is important because emotion plays an important part in purchase decisions. This is where sentiment analysis comes in, giving brands the ability to interpret whether the things reviewers say are positive, neutral or negative.

In a previous article we discussed how text analytics can uncover important themes discussed by reviewers. Sentiment analysis can be applied to these themes, enabling businesses to measure the individual components of their product or service offer. While an overall rating tells you what customers think of the offer as a whole, sentiment analysis reveals what people think about the various parts of it. For example, a retailer might learn that customers are largely positive about their range but are more negative about customer support. Similarly, a hotel on TripAdvisor might focus solely on overall scores, missing negative mentions of specific pain points like noisy air conditioning or a lack of vegan breakfast options. By breaking down sentiment across different themes, businesses gain a clearer picture of what’s driving satisfaction or dissatisfaction.

Revealing trends

Another important aspect of sentiment analysis is its ability to reveal patterns over time. Trends in customer sentiment can highlight shifts in public perception, signalling whether changes to a product or service have hit the mark. For example, if an online retailer introduces faster delivery options, sentiment analysis could quickly reveal whether customers are noticing and appreciating the improvement. This kind of feedback loop, direct and unfiltered, offers a good opportunity to stay in step with consumer expectations.

Benchmarking against the competitive set

Sentiment analysis can also help brands understand how they compare with competitors. By applying the same approach to reviews of rival businesses, brands can uncover the things they are better at than others in the competitive set and the things they need to improve. These insights can inform strategy and sharpen a brand’s competitive edge.

Overall

Sentiment analysis offers more than just data. It helps brands to focus on what’s important, to adapt where necessary and to connect with their audience in meaningful ways.

These connections can make all the difference.

Start Making Sense

Fri, 17 Jan 2025 17:03:53 GMT

How text analytics delivers game-changing business insights

Every day, customers leave behind enormous brand insights in the form of reviews, social media posts and survey responses. Yet most of this data is unstructured, making it challenging to extract meaningful insights. This creates a missed opportunity.

But what if you could use this data to uncover what customers really think not just about your brand but also your competitors, reduce churn and spot emerging trends?

This is where text analytics comes in. Text analytics uses natural language processing techniques to interpret and organise unstructured text. By employing methods like thematic and sentiment analysis, it allows brands to uncover the key themes driving consumer conversations and the emotions behind them.

There are four key uses for text analytics in relation to customer feedback:

1. Understand what customers really think about your product

Traditional surveys and focus groups often provide surface-level insights but can often struggle to capture the depth of consumer feeling. Text analytics dives deeper, revealing how customers actually feel about your products and services. Are they praising your durability but frustrated by usability? Do they appreciate your customer service but feel let down by delivery times?

These granular insights enable teams to refine their offerings with precision, ensuring they address the issues that matter most to their audience.

2. Gain competitive insights to outperform your rivals

Your customers aren’t just talking about you, they’re talking about your competitors too. Text analytics can be used to analyse reviews, social media posts and other feedback to highlight how your brand compares the competition. Are your rivals excelling in areas where you’re falling short? Are they struggling with challenges you’ve already overcome? These insights provide a roadmap for positioning your brand more effectively, allowing you to capitalise on their weaknesses and differentiate yourself in a crowded market.

3. Reduce churn by identifying customer frustrations early

Every piece of customer feedback is an opportunity to retain a customer but only if you know where to look. Text analytics helps identify patterns that can often signal frustration. Whether it’s slow delivery times, confusing return processes or a product that isn’t meeting expectations, these insights enable businesses to act before customers decide to take their loyalty elsewhere. Proactively addressing these pain points can significantly improve retention rates, saving both revenue and reputation.

4. Spot emerging trends before your competitors do

Consumer preferences shift rapidly and staying ahead requires more than just monitoring your own brand. Text analytics detects emerging trends in real time, uncovering what’s gaining traction in your industry. Whether it’s a growing interest in sustainable materials, a preference for convenience over price or an evolving product feature, these insights allow businesses to pivot their strategies early. This proactive approach can mean the difference between being a trendsetter and playing catch-up.

Our approach: Human-Centred machine learning

We combine cutting-edge machine learning with human expertise to ensure that the insights you receive are not just data points but actionable intelligence. This approach ensures precision and relevance, making it easier to translate insights into strategies that drive real results.

In today’s fast-paced market, understanding your customers isn’t optional, it’s the key to survival and growth. Our capabilities gives you a competitive edge by turning unstructured data into actionable insights. Whether it’s reducing churn, spotting trends, enhancing your product or outperforming competitors, the benefits are clear and immediate.

Don’t let valuable customer feedback go to waste. Contact us to see how we can help your business understand the voice of the consumer.

The State of Insight 2025

Thu, 09 Jan 2025 16:19:04 GMT

The State of Play in the Insight Industry: A Call for Change

Overview

The insight industry stands at a crossroads. Once driven by the pursuit of genuine understanding and actionable knowledge, it now finds itself grappling with a host of challenges that threaten the very quality of the work it produces. As someone deeply involved in this space, I believe it’s time to take a hard look at where we are and what needs to change.

The Quality Crisis in Online Surveys

Online surveys have become a staple of market research, but their widespread use has exposed a fundamental flaw: poor-quality samples. Many survey respondents are members of multiple panels, leading to overexposure and diminished reliability. Worse still, we see the rise of "Walter Mitty" respondents—individuals who fabricate answers, either for monetary reward or to pass the time. This behaviour distorts results and undermines the trustworthiness of the data we collect.

The current model for online surveys is unsustainable. A fundamental rethink is needed, starting with how we source respondents and verify the authenticity of their contributions.

Technology can play a role, but no amount of machine learning can replace rigorous validation processes or a commitment to high-quality sampling.

The Over-Reliance on AI

Artificial Intelligence has brought remarkable efficiencies to data collection and analysis, but its limitations are glaringly apparent. While tools like ChatGPT can help speed up analysis, they fall short when it comes to understanding the ‘why’ behind the data. AI cannot yet relate insights to real-world business situations or provide the nuanced interpretations that come from human expertise.

Humans remain essential to the process of generating insight. They bring context, critical thinking, and the ability to probe deeper into anomalies and patterns. AI should be a tool to enhance our work, not a crutch that replaces it. As an industry, we must resist the temptation to hand over too much to the algorithms and ensure that human judgment remains at the centre of our practice.

Synthetic Data: A Mixed Blessing

Synthetic data is a divisive topic. On one hand, it’s essentially fabricated information and risks amplifying underrepresentation of certain demographics. On the other, imputing data to cover gaps is not a new concept; researchers have been doing this for years in more manual ways.

The real question is how we use synthetic data responsibly. It should complement, not replace, real-world data collection and must be deployed with an acute awareness of its limitations. Synthetic data has potential, but it must be handled with care to avoid introducing further biases into our insights.

DIY Survey Platforms: Undermining Expertise

DIY survey platforms have been heralded as a way to democratise research, but in reality, they have legitimised poor-quality practices. These platforms allow anyone to design and launch surveys, regardless of their expertise. The result? Poorly designed studies, biased questions and low-quality data that do more harm than good.

The proliferation of DIY platforms undermines the value of professional researchers and risks normalising mediocrity in insight generation. Expertise matters, and when it is sidelined, businesses suffer from insights that lack depth, accuracy and relevance.

The Decline of Senior Expertise

The rush to cut costs has led to a worrying trend: the loss of senior expertise from both agencies and brands. Experienced professionals are being replaced by junior staff who, while capable, often lack the depth of knowledge required to interpret complex data and translate it into actionable insights.

This cost-cutting approach is shortsighted. Insight is only valuable when it’s paired with strategic thinking and business acumen—qualities that come with experience. Agencies and brands must recognise that investing in senior expertise is not a luxury but a necessity.

What Else Needs Addressing?

Beyond these core issues, there are three other challenges that deserve attention:

The Lack of Standards for Emerging Methods : As new technologies and methodologies emerge, the industry has struggled to establish clear standards and best practices. This lack of consistency undermines trust and comparability.
Data Privacy and Ethical Concerns : With increasing scrutiny on data usage, the industry must prioritise ethical practices and transparency. This is not just a compliance issue but a matter of maintaining public trust.
The Pressure for Instant Results : Businesses are increasingly demanding quick-turnaround insights, often at the expense of depth and accuracy. The industry must push back against this trend and advocate for the value of thorough, thoughtful research.

A Way Forward

The insight industry must take a stand. We need to:

Reassess the role of online surveys and prioritise sample quality.
Use AI as a supporting tool, not a replacement for human expertise.
Develop ethical frameworks for synthetic data.
Reassert the importance of professional researchers and senior expertise.

By addressing these issues head-on, we can ensure that our industry remains relevant, trustworthy, and capable of delivering insights that truly drive business success.

The stakes are too high not to.

That Was The Month That Was

Mon, 06 Jan 2025 16:06:09 GMT

We analyse Google to identify some of the things most searched for in December 2024

For a reminder of the people, politics, news items, events, sports, films and TV programmes searched for in previous months click here .

New Year resolutions from a behaviour change perspective

Thu, 02 Jan 2025 17:03:33 GMT

How capability, opportunity and motivation can boost resolve

New Year, New me

As the New Year dawns, the tradition of making resolutions is as popular as ever. Whether it's shedding a few pounds, starting a new hobby or simply striving to be better, these commitments symbolise hope for transformation. But let’s be honest: how many of us actually stick to them?

Research suggests that 80% have broken their resolution by the middle of February and only 8% stick it out until the end of the year. Why? Because change is hard. To better understand why we falter, it’s worth exploring resolutions through the lens of behaviour change science.

Change isn’t just about deciding to do better. It’s a dynamic process influenced by capability, opportunity and motivation—the three pillars of the COM-B Model. This framework, developed for behaviour change interventions, explains how these components interact to drive or hinder our actions.

Let’s take a common resolution: getting fit.

Capability

Do you have the knowledge, skills, or physical ability to perform the behaviour? For someone starting a fitness routine, this might mean understanding proper techniques, building stamina or learning how to structure a workout. Capability also involves mental resources, such as knowing how to manage time effectively, which can make the difference between success and failure.

Opportunity

Does your environment enable the behaviour? Opportunity can be physical, like having access to a gym or safe outdoor spaces for exercise. It’s also social, encouragement from friends or family or even a culture that supports fitness goals, can create a positive environment for change. On the flip side, obstacles like a busy schedule or lack of facilities can quickly derail progress.

Motivation

Do you genuinely want to make the change? Motivation is often the linchpin of behaviour change, encompassing both reflective drivers (conscious planning and goal-setting) and automatic ones (emotions and habits). For example, setting a goal to feel healthier might be paired with the emotional reward of enjoying activities with friends. However, relying solely on willpower can lead to burnout as motivation needs to be nurtured and supported.

By addressing these three components, the COM-B Model highlights practical ways to overcome barriers to change. For example, someone struggling to maintain a fitness routine might focus on building their capability by starting small, for example with a 10-minute walk each day. They might then create opportunities by blocking time on their calendar for exercise or finding a workout partner for additional motivation. To sustain motivation, they could track their progress or celebrate small wins along the way.

Why do resolutions fail?

So why do so many resolutions fail? Often, it’s because we overlook one or more of these elements. A vague goal like “I want to get fit” doesn’t address capability, opportunity or motivation. Equally, aiming to transform your life overnight is unsustainable without building the right foundations.

Resolutions aren’t about superhuman willpower. They’re about recognising the factors that influence behaviour, creating supportive environments and celebrating progress along the way.

The Worst 10 Christmas Movies

Wed, 18 Dec 2024 14:19:27 GMT

The ones not to watch this Christmas

The Great AI Insight Delusion

Thu, 12 Dec 2024 15:59:51 GMT

Why Generative AI will never replace human insight

Remember when blockchain was supposedly going to revolutionise everything from finance to your morning coffee? Yet here we are, years later, with blockchain quietly sulking in the corner like that gadget you thought you needed but never actually used.

Now, the insights industry has found a new toy to obsess over: Generative AI. Yes, the same technology that can conjure up images of people with six fingers on each of their three arms is now being hailed as the future of data analysis, qualitative research and even report generation.

But let’s be honest: just because AI can do something, doesn’t mean it should.

Generative AI: The emperor's new clothes

Let’s start with the basics. Generative AI, for all its hype, is still in its awkward teenager phase. Yes, it can churn out charts, graphs and even reports at the speed of light but at what cost? These reports are often riddled with errors, lacking in context and dangerously devoid of the nuanced understanding that only human insight can provide. Remember, AI doesn’t “understand” your data. It processes patterns and spits out predictions, often with the finesse of a bull in a china shop.

Imagine handing over your most critical business decisions to an AI-generated report. It might be flashy, it might be fast but like the AI-generated images of human hands, it’s fundamentally flawed. Even if the numbers add up, what do they mean? What’s the story behind them? The AI doesn’t know and it really doesn’t care.

AI v Human insight: There’s no contest

Let’s be honest, the magic of market research isn’t in the data. It’s in the analysis, the interpretation and the ability to draw meaningful insights that drive strategy. This is where human expertise wins hands down. While AI can process vast amounts of data in seconds, it lacks the ability to discern what really matters. It can’t tell you why a trend is happening, how it connects to broader market shifts or what it means for your brand in the long run. That requires human intelligence and no AI can replicate it.

Yes, AI can help with the heavy lifting. It can sift through mountains of data, identify patterns and even suggest correlations. But at the end of the day, it’s still just a tool. It's more like a sophisticated calculator than a crystal ball. The real value comes from the human researchers who take that raw data and turn it into actionable insights. And that’s something AI will never be able to do.

The future of insight: AI as a support, not a replacement

If you're a client and an agency gives you AI generated research, ask for your money back. AI generated reports, charts and analyses will never match the depth and accuracy of human-derived insights.

The real power of AI lies in its ability to support researchers, not replace them. It can help us work faster, dig deeper and explore more data than ever before. But it will never replace the nuanced understanding, creative thinking and strategic insight that human researchers bring to the table.

Let’s not make the same mistake we made with blockchain, elevating a technology to mythical status only to realise it’s not the panacea we hoped for. Instead, let’s use AI for what it is: a powerful tool that, when used by skilled human researchers, can enhance our work and help us uncover insights that truly matter.

In the end, there’s no substitute for human insight. And that’s a fact that no amount of AI-generated nonsense can change.

Bridging The Affordability Gap

Wed, 11 Dec 2024 16:24:21 GMT

The affordability gap is getting wider.

The UK housing affordability crisis has become one of the most pressing socio-economic challenges of recent decades. The scale of the problem is underscored by three key indicators: the steady rise in average house prices, the house price-to-earnings ratio, and mortgage payments for first-time buyers as a percentage of take-home pay.

Recent data from the Nationwide reveals the dramatic rise in average UK house prices over the past decade. Since 2014, average prices have surged from around £190,000 to over £270,000 by late 2024, an increase of nearly 50% in just ten years. Several factors have contributed to this relentless upward trajectory. Record-low interest rates throughout much of the 2010s made borrowing more affordable, fuelling demand and allowing buyers to stretch their budgets. Additionally, the COVID-19 pandemic briefly accelerated house price growth, driven by government incentives such as stamp duty holidays and a surge in demand for larger homes as remote working became widespread.

However, even as interest rates have risen sharply in recent years to combat inflation, average house prices have remained high, underscoring a fundamental imbalance between supply and demand.

House prices have outstripped average earnings providing further evidence of the affordability challenge. This ratio, which compares house prices to average earnings, has shown a stark and sustained increase since the mid-1990s. While the long-term average ratio sits at around 4.5, recent years have seen it remain well above 6, reflecting the extent to which house prices have outpaced wage growth.

The decoupling of house prices from earnings has made it increasingly difficult for prospective buyers to save for deposits or secure mortgages large enough to purchase homes.

A key driver of this trend has been the chronic undersupply of housing, particularly in high-demand areas such as London and the South East. Coupled with rising demand from buy-to-let investors and, in some regions, second home owners and even foreign buyers, this has created a market where prices have risen far faster than incomes for decades.

First time buyers (FTBs) are most significantly affected and many are now faced with increasingly unaffordable mortgage payments.

Historically, first-time buyer mortgage payments as a share of take-home pay fluctuated around a long-term average of around 30%. However, this figure has climbed dramatically in recent years, approaching levels last seen during the housing market peaks of the late 1980s and mid-2000s.

Rising house prices and the recent surge in interest rates have significantly increased the cost of monthly repayments, leaving many new buyers financially stretched. In London and parts of the South East, FTBs are paying over 50% of their take home pay on a mortgage.

For younger generations, who also face challenges such as student debt and stagnant wage growth, this affordability crisis has pushed home ownership further out of reach.

Addressing this housing crisis requires a coordinated and sustained response. Increasing the supply of homes is the most critical step. Successive governments have consistently failed to meet housing targets, particularly for affordable and social housing. Reforming planning regulations, incentivising faster construction and empowering local councils to build more social housing could help alleviate the acute supply shortage. Rising land costs and restrictive planning rules in high-demand areas must also be tackled to enable the construction of more homes where they are most needed.

Labour has pledged that 1.5m new homes will be built over the next five years to tackle the affordability crisis. For context, 997,000 were built in the last 5 years.

Bridging the affordability gap is going to take a monumental effort.

Getting used to inactivity

Fri, 06 Dec 2024 16:05:29 GMT

Does it matter that the number of economic active people is still high?

As at November 2024, 9.25 million people aged 16-64 in the UK were economically inactive (defined as students, people looking after family and home, long-term sick and disabled, temporarily sick and disabled, retired people). This represents 21.8% of this age group.

The inactivity rate was falling steadily until 2020 when it increased during covid. Increases in economic inactivity in the first year of the pandemic were largely driven by those aged 16 to 24 years. Following the pandemic, increases were driven by those aged 50 to 64 years

The latest data from ONS shows that while the rate is slightly down from last year, the level of economic inactivity is still not back to its pre covid levels. This sustained increase presents complex challenges and opportunities for the UK economy. This article explores the underlying causes, the implications for the economic landscape and potential strategies to navigate this new reality.

The Persistence of High Economic Inactivity: Causes and Contributing Factors

The reasons behind the persistently high levels of economic inactivity are multifaceted, reflecting both the direct impact of the pandemic and broader socio-economic trends. Key factors include:

Long-Term Health Issues: The pandemic has exacerbated long-term health conditions and introduced new ones, such as long COVID, increasing the number of individuals unable to work.
Shifts in Retirement Patterns: The economic uncertainty triggered by the pandemic led some individuals to retire earlier than planned, while financial rebounds have made retirement more feasible for others.
Changes in Workforce Demographics: An ageing population means a larger portion of the population is reaching retirement age, contributing to higher inactivity rates.
Evolving Attitudes Towards Work and Life Balance: The pandemic prompted a re-evaluation of priorities, with more individuals choosing to focus on family, education, or personal well-being over traditional employment.

Implications for the Economy

The enduring high levels of economic inactivity in the UK cast a long shadow over the nation's economic landscape. This situation strains the labour market significantly, as a reduced working-age population intensifies existing skills shortages and hampers the country's growth prospects. Furthermore, a larger contingent of economically inactive individuals amplifies the demand for social welfare programmes, putting additional pressure on public finances. This scenario underscores the importance of addressing productivity challenges; with fewer individuals staying in the workforce, enhancing productivity becomes imperative for maintaining economic momentum.

The stakes are high, and the implications of inactivity resonate across the economic spectrum, highlighting the need for strategic interventions.

Navigating the Future: Strategies and Solutions

To navigate the challenges presented by heightened economic inactivity, a comprehensive and nuanced strategy is essential. First and foremost, bolstering health and rehabilitation services will play a critical role. By providing robust support for individuals with long-term health conditions, we can lower the barriers to employment and facilitate their reintegration into the workforce. Additionally, embracing flexible working arrangements stands out as a promising avenue to attract a wider pool of participants into the labour market. Policies that favour work-life balance, including flexible hours and remote work opportunities, can cater to those with caregiving duties or health-related constraints, making the workplace more accessible.

Further, a commitment to lifelong learning and skills development is crucial in an era of rapid technological advancement and economic change. By fostering an environment that encourages continuous education and training, individuals can acquire the skills necessary to thrive in the evolving job market. Targeted support for older workers, such as phased retirement options and retraining programs, can also mitigate the impact of an ageing population, encouraging them to remain in or return to the workforce.

Lastly, fostering a culture of inclusivity within the workplace can inspire a broader spectrum of the population to engage in economic activities. An inclusive environment, which accommodates diverse needs and backgrounds, is vital for drawing more people into the economy. This approach not only addresses the immediate challenges of economic inactivity but also lays the groundwork for a more resilient and inclusive economic future. Through collaborative efforts across government, industry, and community sectors, the UK can rise to meet these challenges, paving the way for sustainable economic growth and prosperity.

Conclusion

The landscape of economic inactivity in the UK has shifted significantly, with a complex array of factors contributing to current levels. While the challenges are substantial, they also present an opportunity to reimagine how we approach work, education, and economic participation. By adopting a comprehensive strategy that addresses the root causes of economic inactivity and leverages the potential of an underutilised segment of the population, the UK can navigate these changes and foster a more resilient and inclusive economy.

The journey ahead requires collaboration across government, industry, and communities to create a more adaptable and sustainable economic future.

Travel Beyond The Stars

Wed, 27 Nov 2024 12:25:03 GMT

Unlocking insight from review sites

Customer reviews have become a cornerstone of decision-making in the travel industry. From hotels to cruise liners, leisure providers to bespoke adventure holiday agents, businesses are shaped by what their customers say about them online. Platforms like Tripadvisor provide an endless stream of feedback, but the challenge lies in turning that flood of unstructured data into meaningful insight.

This is where text analytics comes in, offering a powerful way to decode customer sentiment and behaviour to drive business growth.

Rather than simply relying on star ratings or review counts, text analytics digs deeper. It identifies patterns in language, extracts recurring themes, and gauges the emotional tone of reviews. This allows businesses to understand not just what customers think, but why they think it. For example, guests might repeatedly mention “friendly staff” or “clean rooms,” but they may also voice frustration about issues like booking difficulties or crowded facilities. Text analytics quantifies these insights, helping businesses pinpoint what they’re doing well and where they need to improve.

One of the greatest advantages of text analytics is its ability to uncover hidden trends and emerging themes. These might include growing interest in eco-friendly options, demand for more personalised adventure packages, or frustration over outdated amenities. By identifying such trends early, businesses can adapt quickly to meet changing expectations, keeping them competitive in an ever-evolving market.

Understanding the competition is another area where text analytics excels. Customers often compare brands in their reviews, mentioning alternatives they’ve considered or experiences they’ve preferred elsewhere. Analysing this data can highlight competitive advantages or gaps in the market, as well as opportunities for partnerships. For example, a resort might discover that many of its guests are also reviewing local adventure providers, opening the door to collaborative marketing or bundled offers.

Customer reviews also reveal what drives loyalty. Text analytics can help identify the attributes that make travellers return—whether it’s exceptional service, unique experiences, or seamless logistics. These insights can shape both operational improvements and marketing strategies. Instead of relying on assumptions about what customers value most, businesses can ground their decisions in data, ensuring their investments align with real customer priorities.

At its core, text analytics transforms raw customer feedback into a strategic resource. It equips travel businesses with the tools to enhance customer satisfaction, refine their offerings, and stay ahead of industry trends. In a world where word of mouth increasingly takes place online, understanding and acting on what customers say has never been more critical.

With text analytics, businesses can do more than listen to their customers - they can truly understand them, creating experiences that leave a lasting impression and inspire future bookings.

Cost Effective Segmentation

Wed, 27 Nov 2024 11:16:45 GMT

Using existing data to deep dive into your audience

Understanding your audience is essential for building a strong brand. Dividing your audience into distinct groups or segments based on shared characteristics, helps tailor products, services and communications to meet their needs. However, traditional segmentation studies can be expensive as they require bespoke research and extensive data collection.

The good news is that brands don’t always need to spend heavily with research agencies to gain meaningful insights. By using data already at their disposal and combining it with publicly available and geo-demographic data, businesses can create effective segmentations at a fraction of the cost.

Using your own business data as a foundation

Every business generates valuable data through its daily operations. Customer Relationship Management (CRM) systems, for example, contain valuable information about customer behaviour. Purchase histories, customer demographics and engagement patterns can all reveal insights into distinct audience groups. By analysing this data, businesses can identify segments based on factors like buying frequency, average spend and product preferences.

Sales data can also provide clarity. Understanding which products or services are most popular in specific locations, among certain age groups, or at different times of the year can highlight trends and clusters within your customer base. Similarly, customer feedback, complaints and inquiries offer a qualitative dimension, helping to identify not just what customers buy, but why they buy it (or why they don’t).

Using publicly available data to build on the foundations of your segmentation

Public data sources can enhance your segmentation efforts without adding significant costs. Social media platforms can provide insights into customer interests, hobbies and even purchasing behaviours through comments, shares and interactions with branded content. Review sites like Trustpilot can shed light on customer preferences and pain points. These platforms also allow brands to observe competitors’ customers, offering a comparative perspective on their audience.

Market reports and industry studies, often published freely or at a low cost by trade associations or government bodies, can provide valuable demographic and psychographic insights. Tools like Google Trends and social listening platforms can further highlight broader consumer trends and emerging topics relevant to your brand’s audience.

Adding geo-demographic data is another powerful yet cost-effective tool. This data links geographic locations with demographic and lifestyle characteristics, allowing businesses to understand how different segments cluster by area. Census data or commercially available tools like Experian’s Mosaic provide detailed profiles of neighbourhoods, helping brands tailor marketing strategies to fit local needs. For example, a brand might discover that its most loyal customers live in suburban areas with young families. Using geo-demographic data, it could identify similar areas in new markets for targeted advertising. Alternatively, combining this data with sales patterns might reveal underserved regions or hidden growth opportunities.

Creating a cohesive picture

The true power of these data sources lies in their integration. By combining CRM and sales data with public and geo-demographic information, brands can create a multi-dimensional view of their audience. For example, a retail company might segment customers based on both spending habits (internal data) and lifestyle preferences (public and geo-demographic data).

This hybrid approach ensures segments are not only actionable but also rooted in real-world behaviours and trends.

Advances in analytics and visualisation tools make this integration more accessible than ever.

A practical, budget friendly approach

Cost-effective segmentation doesn’t mean cutting corners; it means working smarter with the resources already available. By focusing on internal, public and geo-demographic data, brands can develop actionable audience profiles without the financial burden of custom research. This approach not only reduces costs but also ensures segmentations are grounded in the realities of your business, making them more relevant and effective.

In today’s data-rich world, understanding your audience doesn’t have to break the bank. By making the most of what’s already at your fingertips, you can unlock the full potential of your customer data and position your brand to connect with its audience in meaningful ways.

Using Data to Optimise Marketing ROI

Wed, 20 Nov 2024 10:54:59 GMT

Insight that adds value to marketing campaigns

In today’s fragmented media landscape, brands are spoiled for choice when it comes to reaching their audiences. From the broad reach of TV to the immediacy of digital, each channel offers unique strengths. However, this variety also presents a challenge: ensuring that a multi-channel marketing campaign delivers a strong return on investment (ROI).

The secret lies in insight, understanding not just how each channel performs but how to use their collective strengths to engage audiences and drive outcomes.

The role of insight in campaign planning

Before a campaign even begins, insight shapes its direction. Understanding your audience is fundamental: who they are, where they spend their time and what resonates with them. This is where data from past campaigns, customer surveys and broader market research come into play. For example, demographic data might reveal that your target audience consumes TV differently depending on age or region, while social listening could uncover trending topics or cultural moments you can tap into.

Equally important is understanding how channels interact. TV might deliver broad awareness but digital channels often provide the means for immediate interaction or conversion. Radio’s ability to reach audiences during commutes or downtime complements the visual impact of print and the immediacy of mobile ads. By aligning these channels around a cohesive message, brands can amplify their impact.

Engaging audiences in the execution phase

When the campaign is live, insight is your real-time feedback loop. Each channel offers distinct signals about what’s working and what isn’t. For example, TV campaigns might not offer immediate feedback but can be tracked through uplift in brand searches or direct traffic. Radio and print, while traditionally harder to measure, benefit from tools like unique URLs, QR codes that provide tangible data points on effectiveness.

In contrast, digital channels deliver a near-constant stream of performance metrics. Click-through rates, conversions and engagement levels provide granular detail about audience behaviour. But the key to leveraging these insights effectively is connecting the dots. A spike in web traffic after a TV ad airs or increased social media mentions during a radio spot suggests that your channels are working together to create a cohesive story.

Adapting and optimising on the fly

The beauty of multi-channel campaigns lies in their flexibility. While certain channels, like TV and print, might require longer lead times and fixed schedules, others, especially digital, allow for real-time adjustments. If a particular creative is underperforming or a new trend emerges, digital platforms offer the agility to tweak messaging, adjust targeting, or increase spend in high-performing areas.

For example, if early data shows that a campaign resonates strongly with a particular demographic, brands can reallocate budgets to focus on that group. Similarly, underperforming elements, whether an uninspiring headline or a misaligned call-to-action, can be quickly replaced. This adaptability ensures that even when campaigns don’t land perfectly at first, they can evolve to maximise impact.

Post-campaign learnings for long-term success

Once the campaign ends, the real value of insight becomes clear. Evaluating performance across channels is not just about measuring success but also about identifying what worked, what didn’t and why. Advanced tools like multi-touch attribution and econometrics help brands understand the contribution of each channel to overall ROI. For instance, how did the emotional resonance of a TV ad drive online searches? Or how did a digital retargeting strategy extend the reach of a print ad?

But post-campaign analysis isn’t just about hard metrics.

Qualitative insights, such as customer feedback, social media sentiment or even anecdotal evidence from sales teams, offer valuable context. Perhaps a specific message struck a chord because it aligned with a broader cultural moment, or maybe the campaign succeeded in sparking conversations in unexpected ways.

These learnings are the building blocks for future campaigns.

The art of integration

Ultimately, the most successful campaigns aren’t those that focus on individual channel performance but those that understand how channels work together. Insight helps brands break down silos and create a cohesive strategy that leverages the unique strengths of each medium. A TV ad might capture attention, radio builds familiarity through repetition, digital provides interactivity and print reinforces credibility. Together, they create a seamless brand experience that guides the audience from awareness to action.

The challenge for brands is to ensure that insights from each channel don’t remain in isolation. Tools like unified analytics platforms and cross-channel dashboards help paint a holistic picture of campaign performance. By understanding the interplay between channels, brands can uncover opportunities to refine their messaging, improve targeting, and achieve greater efficiency in their spend.

Conclusion

In multi-channel marketing, insight isn’t just a tool; it’s a strategy. It informs planning, guides execution, enables optimisation and drives long-term learning. By embracing an insight-led approach, brands can navigate the complexities of modern marketing with confidence, ensuring that their campaigns not only reach the right people but also deliver meaningful results.

The media landscape may be more fragmented than ever but with the right insights, brands can turn that complexity into their greatest advantage.

Getting all your insight ducks in a row

Wed, 13 Nov 2024 17:03:19 GMT

Aligning insight with strategy

In today’s data-driven world, businesses are overwhelmed with information. From customer behaviour metrics to market trends, the sheer volume of data has transformed how organisations operate. However, the real value of data lies not in its abundance but in its application.

By aligning data insights with business strategy, organisations can move beyond reactive decisions to proactively position themselves to anticipate change, seize opportunities and improve performance.

Why Alignment Matters

Aligning data insights with business strategy ensures that data does not operate in isolation. It bridges the gap between analytics and actionable outcomes, providing a direct path from information to informed decision-making. Without this alignment, even the most advanced analytics can fail to deliver meaningful results.

For example, a retail company may discover that a certain demographic is underrepresented in its sales. If this insight is not connected to a clear strategy, such as targeted marketing or product diversification, it becomes a missed opportunity. Alignment ensures that insights lead to meaningful action.

It also enables businesses to prioritise effectively. Organisations often face competing goals, and data-driven insights help leaders allocate resources where they are most needed. For example, a manufacturing firm may find that supply chain delays are a major contributor to revenue loss. By aligning this insight with its strategic goals, the company can focus investment on strengthening its logistics operations.

Key Benefits of Alignment

When data insights are fully integrated with business strategy, the benefits are significant:

1. Enhanced Decision-Making
Leaders are able to make better, more informed decisions based on current trends and facts rather than outdated assumptions or gut instinct. For instance, a bank could use predictive analytics to identify customers at risk of leaving and introduce retention strategies that directly impact profitability.

2. Improved Agility
Businesses that align data with strategy can adapt more quickly to market changes. For instance, a clothing retailer using real-time sales data can adjust inventory and marketing efforts to remain competitive during seasonal changes or shifts in demand.

3. Maximised ROI on Data Investments
Data initiatives can be costly, but alignment ensures they deliver value. By focusing on strategic priorities, organisations avoid wasting resources on collecting or analysing irrelevant data.

4. Strengthened Customer Relationships
Understanding and acting on customer data allows organisations to personalise experiences, improving loyalty and lifetime value. For example, a SaaS provider might use usage data to recommend new features or upgrades tailored to individual clients.

5. Cross-Functional Collaboration
Aligning insights with strategy requires cooperation across departments. Breaking down silos between marketing, operations, finance and sales ensures that data insights are interpreted and implemented collaboratively.

Real-World Applications

The alignment of data insights with business strategy takes different forms in various industries, but the principles are consistent:

E-Commerce and Personalisation
Businesses like Amazon and ASOS excel at aligning data with their strategies. By using browsing and purchasing data, they create personalised recommendations, targeted campaigns and dynamic pricing strategies that enhance customer satisfaction and retention.

Healthcare and Operational Efficiency
Hospitals analysing patient flow data can identify and address bottlenecks in emergency departments, ensuring resources are deployed effectively and improving patient outcomes.

Financial Services and Risk Management
Banks use data to align risk management strategies with regulatory requirements. By analysing transaction patterns, they can detect fraud early and implement measures to protect both customers and operations.

Sports Teams and Fan Engagement
Football clubs increasingly use data to align fan engagement strategies with their commercial goals. By examining social media interactions and ticket purchase patterns, they create campaigns that connect with fans and drive merchandise sales and attendance.

Making Alignment Work

To successfully align data insights with business strategy, organisations should take a structured approach:

Define Strategic Goals
Clear objectives are essential to ensure data insights are relevant and actionable.
Invest in Tools and Talent
Effective alignment requires reliable analytics tools and skilled professionals who can interpret data and link it to business outcomes.
Foster a Data-Driven Culture
Leaders must champion the use of data in decision-making and encourage teams to align insights with strategic goals.
Monitor and Adjust
Alignment is an ongoing process. Organisations should regularly review and refine their strategies as new data becomes available.

Conclusion

In today’s competitive environment, organisations cannot afford to treat data as an afterthought. Aligning data insights with business strategy ensures that every decision is based on evidence and supports overall objectives. From improving operational efficiency to enhancing customer engagement, the benefits are wide-ranging. As technology continues to advance and data grows in importance, businesses that achieve this alignment will be best placed for future success.

Gut Health

Fri, 08 Nov 2024 16:10:09 GMT

Getting a healthy balance between gut instinct and data driven decision

You're in charge. You've got 25 years experience. You know what you're doing. You go with your gut because you've never been wrong yet.

This scenario plays up and the down the country every day. And for many senior leaders it works for them and their business. But as organisations become more data-rich, many senior leaders find themselves at a crossroads: to rely on instinct, honed by years of experience or to turn to data, which promises objectivity and predictive power. Yet, rather than choosing one approach over the other, the most effective path may lie in harmonising these two methods to create a more balanced, adaptive, and resilient decision-making process.

The Case for Data-Driven Decision-Making

Data-driven decision-making has transformed business operations, creating what many would consider an objective and rational foundation for strategy. Organisations can now harness insights from a variety of sources—consumer behaviours, sales trends, market analytics, and more—to guide their choices. This approach promises precision and mitigates the potential for personal biases, which can cloud judgment. For senior decision-makers, data offers the ability to substantiate business strategies and predict future scenarios with increased accuracy, providing a sense of certainty that can be particularly reassuring in uncertain times.

One of the most significant benefits of data-driven decisions is their scalability. Unlike intuition, which can vary wildly between individuals, data provides a common language across an organisation. Key performance indicators (KPIs), conversion rates, and customer retention metrics give everyone—from the boardroom to the front lines—a shared, measurable standard of performance. In sectors such as retail or finance, where variables change rapidly and small changes can make a significant impact, data offers a valuable guide for recalibration and course correction.

However, relying solely on data can have limitations. Data can sometimes lack the nuance of real-world experience, failing to account for unforeseen variables or qualitative factors that numbers simply cannot capture. A data-led approach can also suffer from “paralysis by analysis,” where the sheer volume of information leads to overthinking and indecisiveness. Data, while powerful, is rarely perfect. It can be incomplete, biased, or even misleading, especially when over-reliant on historical performance rather than forecasting future conditions. For many senior leaders, trusting data alone can feel restrictive, as it lacks the dynamic, real-time responsiveness that characterises high-stakes business environments.

The Value of Intuition in Decision-Making

At the opposite end of the spectrum, intuition—often referred to as “gut instinct”—plays a fundamental role in decision-making, especially for seasoned leaders. This intangible “sixth sense” is born out of years of experience, a deep understanding of the business landscape, and an awareness of subtleties that data might not capture. Intuition allows leaders to act decisively and swiftly, particularly in situations where data is lacking or ambiguous, and provides a compass in the face of uncertain, high-pressure situations.

For example, a CEO may decide to pivot a product line or enter a new market not based solely on a spreadsheet, but because of an intrinsic sense of the opportunity that arises from years of navigating similar situations. This kind of intuitive decision-making allows organisations to act faster, often necessary in highly competitive markets. Furthermore, intuition can be a powerful differentiator, driving creative and innovative thinking that isn’t tethered to historical data.

Yet intuition alone is not without risks. It can be highly subjective, and while experience informs it, so do biases and emotions. Over-reliance on intuition can lead to inconsistency and misjudgement, particularly if a leader’s personal views cloud objectivity. What worked in the past may not necessarily apply to a current situation, especially in a rapidly evolving business landscape. As such, intuition needs to be tempered with checks and balances, especially when stakes are high.

The Best of Both Worlds: A Blended Approach

So, how can senior decision-makers find the right balance between these two approaches? Increasingly, the answer lies in a blended model that combines data-driven insights with gut instinct. This approach acknowledges that while data is critical in providing evidence-based insights, intuition serves as an equally important guide, offering the flexibility and responsiveness that pure data analysis might lack.

A balanced approach allows leaders to leverage the best of both worlds. Data can offer a factual foundation, helping to validate or challenge gut instincts, and enabling leaders to make informed decisions with greater confidence. Conversely, intuition can act as a filter through which data is interpreted, allowing leaders to weigh the information against their own understanding and experience. Rather than letting one approach override the other, a blend creates a dynamic decision-making model that is informed yet agile.

For example, when evaluating the potential for a new product launch, a data-driven model might offer insights into market demand, customer demographics, and competitive positioning. But it’s often intuition that provides the “human” element, sparking questions that data alone might not raise: Are we confident in this product’s long-term appeal? How will this resonate with our core customer base? What might be the unintended consequences? Together, data and intuition create a more holistic view of the decision at hand.

Building a Data-Intuition Culture

To foster this balance, organisations should build a culture that values both data and intuition, encouraging collaboration between data analysts, strategists, and decision-makers. Training leaders to interpret data without losing sight of instinct can help bridge the gap between analysis and action. Moreover, investing in high-quality, accessible data sources and tools can empower senior leaders, giving them the information they need to feel confident in data while keeping room for their intuition.

In a world where data is increasingly available, it may seem tempting to let it be the sole guide. However, data alone can’t capture the complexities and unpredictabilities of the market. Combining it with intuition means recognising that every number has a context and that experience provides insights that raw data cannot.

Ultimately, the best business decisions are neither purely data-driven nor entirely intuitive. Instead, they lie at the intersection of facts and experience, rationality and instinct. By combining data with intuition, leaders can navigate ambiguity with confidence, making decisions that are not only well-informed but also resilient, adaptive, and in tune with both the science and the art of business.

Halloween movie series

183:752711313 (Michael Wagstaff) — Wed, 30 Oct 2024 10:31:56 GMT

1978 original tops worldwide box office returns

Publicly Available Metrics That Add Value

Thu, 24 Oct 2024 16:04:09 GMT

The Five You Can't Afford To Miss

Recently we identified the ' magnificent' seven metrics that should be tracked in a market research brand tracker. To build a well-rounded understanding of your brand’s perception and position in the market, traditional brand-tracking surveys should be complemented with publicly available insights. These insights, derived from online sources like reviews and social media, can help you identify strengths, weaknesses and opportunities for growth.

Below are the five key metrics that will deepen your understanding of brand sentiment, competitive positioning and potential partnerships.

Sentiment Analysis Across Review Platforms

While volume of reviews is widely monitored, it’s crucial to dig deeper by analysing review sentiment—both quantitative ratings and qualitative comments across major platforms like Trustpilot, Amazon and Yelp. Tracking the overall positive or negative sentiment of your reviews over time provides insight into whether your brand experience is meeting, exceeding or falling short of customer expectations.

Why it matters:
By focusing on the sentiment trends, you’ll gain a clear picture of shifts in customer satisfaction and can identify specific areas requiring attention or potential risks to brand loyalty.

Reviewer Mentions of Competitors and Substitute Brands

A unique insight can be found in tracking the brands that your reviewers also mention in their feedback, either as competitors or alternatives. If your brand appears alongside certain competitors, you’re able to map your competitive set more accurately and monitor shifts in their relevance. Tracking mentions of complementary brands can even reveal partnership opportunities.

Why it matters:
Understanding your brand’s place in the market goes beyond self-assessment; it means staying aware of where you sit in the broader landscape. Identifying frequently mentioned complementary brands might also unlock potential alliances to enhance your market positioning.

Customer Pain Points and Product Gaps

Review sites and forums often capture specific pain points and unmet needs, creating valuable insights for refining or innovating products. By analysing recurring issues and product gaps in customer feedback, you’ll uncover potential areas for improvement and innovation, including design flaws, quality issues and feature requests.

Why it matters:
This granular level of insight helps prioritise product developments or fixes, ensuring that your brand evolves in ways that meet customer needs and expectations.

Engagement Quality on Visual Social Platforms

Social media platforms like Instagram and Pinterest provide brand-specific data points beyond just likes or comments. Instead, focus on measuring engagement quality—such as the depth of customer interactions or the sharing of brand-related content. Qualitative interactions, such as thoughtful comments, user-generated posts or story shares, reflect stronger brand resonance than casual likes.

Why it matters:
Meaningful interactions reveal how well your brand’s narrative resonates with your audience, indicating emotional engagement and loyalty that extend beyond passive followers.

Top Review Keywords and Themes

Keyword and theme analysis across reviews and social media comments highlights core perceptions and recurring associations with your brand. Commonly used descriptors and phrases such as “value for money,” “durable,” or “excellent customer service”, sheds light on the most significant brand attributes in the eyes of consumers. By examining positive and negative themes, you’ll understand the nuances of your brand image.

Why it matters:
Identifying dominant themes in customer feedback allows you to assess whether your brand is known for its intended attributes and provides actionable insights for refining your messaging and customer experience.

In Summary

With the right mix of these publicly sourced metrics, you can enrich your understanding of brand health and competitive position in real-time. These insights bridge the gap between brand tracking surveys and on-the-ground perceptions, enabling you to adjust your strategy in a data-driven way.

If you’re looking to dive deeper into optimising brand tracking across both survey and publicly available data, contact us for a complimentary consultation.

The Magnificent Seven

183:752711313 (Michael Wagstaff) — Wed, 23 Oct 2024 15:41:24 GMT

The questions you should ask in your brand tracker

In October we gave advice on how to trim your research costs and enhance its value for money. One of our suggestions was to ditch the brand tracker or cut it right back to a maximum of 10 questions.

We recommend that your 10 questions include what we call the 'magnificent 7' listed below. You can use the remaining three slots to follow up on these or have two or three questions on metrics that are important to your business.

1. Unprompted brand awareness

It's important to know how many people name your brand when asked to spontaneously name brands that operate in your market. Unprompted or spontaneous brand awareness measures the number of people who express knowledge of a brand or product without prompting. This measure is also known as brand recall or share of mind.

A development of unprompted awareness is to ask what brands come to mind on an occasion where your brand would be bought. For example, what brands come to mind when you have friends coming over to watch the football and you need to get in some vegan snacks. Known as brand salience, this type of question will help you to gauge where you are in the hierarchy of memory generated brands recalled in buying situations. The more frequently you are mentioned, the better.

2. Prompted awareness

Unless your brand is a household name then spontaneous awareness is likely to be low which means it is going to prove tricky to ask follow up questions about the brand to a limited audience. This is where prompted awareness comes in.

Showing respondents a list of brands and asking them to select the ones they have heard of is a measure of brand recognition. Used together with unprompted awareness you can begin to see the impact of any marketing campaign to raise awareness particularly among the target market.

3. Perceptions and brand attributes

The perceptions that people have of your brand and the attributes they assign to it are important factors in whether or not they recall it from the memory bank in buying situations. Your tracker should test a number of statements, values and perceptions to see which ones they ascribe to your brand and which ones they ascribe to the competitive set.

There should be a balance between negative and positive perceptions and attributes so that you know which ones to work on. Measuring the speed and strength of responses will reveal just how embedded your brand is in people's minds.

4. Category use and product use

It's important to find out whether respondents already buy the category, what products they buy and how frequently they do so. This will help you to understand your place in the market and the potential size of the prize available by taking market share from rivals.

5. Consideration and intent

There is often a wide disconnect between what people say they do and what they actually do. Nowhere is this truer than in stated intention to buy a product or use a service. That said, asking people how likely they are to buy your product or use your service is a useful indicator of your place in the market and how this can change over time. Just don't make any revenue projections based on it.

6. Spontaneous campaign awareness

Does your target audience recall or recognise any elements of your marketing campaign over a recent time period. Ideally you want people to describe what they saw and where they saw it. Ideally (again) this should be unprompted so that you can measure just how much of an impact on people's mindsets the campaign has had.

7. Recommendation

Some marketers swear by their Net Promotor Score, others swear at it. We are not fans because it does not reflect real life conversations. When I meet up with family and friends the conversation never gets around to whether I would recommend Microsoft products but that might just be me. That said, it's important to assess customer experience with the brand and whether people will stay with it.

The precise wording of the questions will be dependent upon what it is you're tracking and the way the questions have been asked previously. Our magnificent 7, however, will give you an excellent steer on what you should be asking. And of course, you can always contact us for free advice on specific wording .

Telling The Story With Data

Fri, 18 Oct 2024 15:21:18 GMT

Handy tips to find those hidden gems of business insight

In the world of data and insights, the difference between impactful research and a missed opportunity often comes down to how well the data is translated into a compelling story. For senior decision makers, it's crucial that these stories not only inform but drive action.

But how do you decide what the story should be, what metrics to include, and how to ensure that the insights are applied to real business decisions?

We discuss these questions below.

Deciding What the Story Is

The first step in crafting a data story is deciding which story to tell. The goal is not to summarise all the data available, but to focus on what truly matters to the business at this moment.

To determine the story:

Start with Business Objectives: What is the business trying to achieve? Are you looking to improve customer loyalty, grow market share or fine-tune product positioning? The story should always align with the strategic goals of the organisation.
Identify Key Trends or Anomalies: Review the data to spot significant patterns, trends or outliers that relate to these objectives. For example, it might be a sudden shift in customer preferences or could be a surge in negative sentiment around a specific product feature.
Ask ‘Why’ and ‘So What’: Every story should answer not just what is happening but why it matters to the business. If the data shows a decline in customer satisfaction, explore the underlying reasons and link them to broader business implications.

The most compelling stories come from a balance of understanding the data and connecting it to the bigger picture that resonates with decision makers.

Choosing the Right Metrics

Choosing the right metrics is crucial for telling a story that is both compelling and actionable. Not every data point needs to be included; only those that contribute meaningfully to the narrative.

Here’s a guide to help select them:

Focus on Key Performance Indicators (KPIs): Start with the metrics that are directly tied to the business’s KPIs. These are the numbers that decision makers care most about because they reflect business health, for example customer retention rates, net promoter scores or sales growth.
Use Diagnostic Metrics for Depth: KPIs alone don’t always tell the whole story. Include diagnostic metrics to explain why a KPI has changed. For example, you might pair sentiment or satisfaction scores with associated brand attributes or key themes to explain why a product has received more negative reviews.
Avoid Data Overload: Senior leaders don’t need to see every number. Be selective and ensure that the data you include directly supports the story. Too much detail can obscure the key message and dilute the impact of your insights.
Contextualise the Metrics: Numbers are more powerful when framed within a context. Benchmarks, historical data or industry comparisons provide the context needed to make the metrics meaningful. For instance, if your market research shows a 10% dip in customer satisfaction, how does that compare with competitors or to the previous year’s results?

Ensuring the Story Drives Action

Once you have the story and the right metrics, the next challenge is ensuring the insights are actually used to drive business decisions.

Here are some pointers to making sure your story leads to action:

Involve Stakeholders Early: Get input from key decision-makers early in the process. Understanding their priorities and concerns helps shape a story that’s relevant to them. This makes it more likely that they will engage with and act on the insights.
Tie Insights to Business Impact: Always connect the story to tangible business outcomes. Rather than just presenting data, explain how taking action on these insights will affect the business. For example, “Improving our customer support experience based on this feedback could boost retention by X%,” provides a compelling reason to act.
Create Clear Next Steps: Make sure your data story ends with actionable recommendations. These steps should be specific, achievable and directly linked to the insights. This gives decision makers a clear path forward, making it easier to turn insights into strategies.
Follow Up on Implementation: Ensure that the insights don’t just sit in a report. Schedule follow-ups to track whether the recommended actions are being implemented and assess their impact. This reinforces the value of data-driven storytelling and strengthens its role in future decision-making processes.

Conclusion

Storytelling with data is more than just a reporting exercise; it’s a strategic tool for driving business decisions. By carefully choosing the story, selecting the right metrics and ensuring actionable insights, data can become the cornerstone of business strategy.

For senior decision makers in brands, it’s not just about what the data says—it’s about how that story drives meaningful, measurable change.

Get in touch if you'd like some free advice about telling your story.

Use Case For Text Analytics

Sun, 13 Oct 2024 15:01:26 GMT

Making business sense of what people say.

Text analytics is the process of extracting meaningful insights from human-written texts, using techniques such as natural language processing to identify patterns, trends, and sentiments. It's an important tool for extracting value from the vast amounts of unstructured data about brands.

This technique helps turn customer feedback, user reviews, verbatim comments on surveys and social media interactions into insights that can shape strategy, refine messaging and improve the customer experience.

In this article, we explore some of the use key use cases where text analytics can be employed to provide insight and discuss why brands should be utilising this capability.

Analysing online customer feedback

One of the most common applications of text analytics is analysing customer feedback. There are a lot of opportunities for customers to share their thoughts and experiences with brands. However, much of this feedback comes in the form of open-ended responses in online reviews.

Text analytics enables us to make sense of this information by identifying common themes, sentiments and the attributes customers associate with brands. This goes beyond simple keyword tracking and allows a deeper understanding of the issues driving customer satisfaction or dissatisfaction. Text analytics can quickly spot emerging trends, whether it’s a recurring complaint about a product feature or praise for a particular aspect of service. By acting on these insights, brands can refine their offerings, improve customer experience and build stronger relationships with their customers.

Customer experience

Text analytics plays a significant role in extracting meaning from the vast amount of unstructured data generated by customer experience channels. Whether the data comes from surveys, call centre transcripts, online chat or open-ended NPS questions, text analytics can aggregate and analyse this feedback quickly.

We use the analysis to help brands prioritise the issues most important to the customer, identify gaps in service delivery and adjust strategies accordingly. For example, if a retailer notices that third party delivery times are frequently mentioned in customer reviews, they can take immediate action to improve this aspect of the customer experience.

The value of text analytics within customer experience programmes lies in its ability to continuously inform business decisions, helping brands stay customer-centric and responsive.

Product development

Product development is another area where text analytics is a valuable tool. Text analytics aggregates and analyses product specific comments to help product teams identify the most common pain points, bugs and potential areas for improvement. For example, if a recurring issue with a product feature is flagged through customer reviews, text analytics can quantify the extent of the problem and provide evidence to support its prioritisation in the development roadmap. This capability not only speeds up the product improvement cycle but also ensures that innovations are aligned with customer needs, improving overall satisfaction and reducing the risk of launching products that fail to meet market expectations.

Text analytics can also be used to monitor rival brands and identify the strengths and weaknesses of products directly against competitors. It uncovers gaps in brand offerings, gain insights into emerging trends and highlights where brands can differentiate or improve their products to better meet market demand.

Reputation management

Reputation management has become increasingly complex as more consumers take to digital platforms to share their opinions. In this environment, the ability to monitor and manage a brand’s reputation is critical. Text analytics allows brands to track online discussions, reviews and media coverage, providing a clear view of how their brand is being perceived. This continuous monitoring helps brands identify potential issues before they escalate into full-blown crises. For example, if negative sentiment around a brand spikes following a controversial ad campaign, text analytics can pinpoint the specific issues that are fuelling dissatisfaction, enabling the brand to issue targeted responses.

In a competitive landscape, maintaining a positive brand image is crucial, and text analytics offers a proactive approach to ensuring that a brand’s reputation is managed effectively.

Conclusion

In conclusion, text analytics offers brands an important opportunity to make sense of the vast amounts of unstructured data they encounter daily. From improving the customer experience to enhancing and supporting product development and brand reputation, the value of these insights is immense.

Get in touch if you're interested in finding out more about how text analytics can help your brand.

Shifting Insights : From Reactive to Proactive

Thu, 03 Oct 2024 15:14:51 GMT

How things have change 40 years on

When I began my journey in the market research industry 40 years ago, we were largely focused on looking back. Our job was to gather data, interpret it and explain what had already happened. We would run surveys, conduct interviews, and facilitate focus groups, collecting responses that described past behaviours, preferences, and attitudes. This data was static and often retrospective. We analysed it diligently, drawing insights from neatly structured datasets to explain why things had happened the way they had.

Fast forward to today, and the world of insights has evolved dramatically. Where once our focus was squarely on the past, we are now increasingly tasked with predicting the future.

A transformative shift

Modern insight generation has shifted from being reactive to proactive. Today, we look forward, using the wealth of data at our disposal not just to explain what has occurred but to anticipate what is coming next. This shift has been driven by several significant changes in the tools, techniques, and data streams we now rely on.

One of the most transformative developments has been the explosion of data. In the early days, we dealt with relatively limited datasets. These were often confined to consumer responses to survey questions, and our challenge was to interpret those responses within the boundaries of a few key variables. The data was usually structured, neatly organised into rows and columns.

But now, data pours in from an ever-expanding range of sources. We no longer rely solely on what consumers tell us; we observe what they do. From transactional data and social media behaviour to geolocation data and even interactions captured by smart devices, we are surrounded by a constant flow of information.

This vast influx of data has enabled us to move from explaining past behaviour to forecasting future trends.

Old meets new

Traditionally, we relied on well-established statistical techniques like regression analysis and predictive modelling to uncover patterns in historical data. These techniques are still at the heart of what we do but today, we enhance them by integrating machine learning to get the best of both worlds – the rigour of proven statistical models and the adaptability of AI-driven insights.

Predictive models allow us to make sense of complex patterns across multiple datasets, helping brands identify trends that might not be immediately apparent. For example, a marketing team that once relied solely on past sales data to forecast demand can now build models that incorporate social media sentiment, user reviews and CRM data, allowing for a more dynamic and accurate prediction of consumer behaviour.

Let’s take a couple of real world examples to explain what we mean: imagine you’re launching a new product. Traditionally, you would forecast sales based on historical data from similar products, perhaps factoring in seasonality or economic conditions. While useful, this approach can miss emerging signals. By blending predictive modelling with machine learning, you can feed real-time data, such as how consumers are interacting with your brand on social media or shifts in competitor behaviour, into your model. As a result, your forecasts become more nuanced and responsive, helping you allocate marketing spend more effectively and optimise inventory levels.

Another example is customer segmentation. Previously, segmentation was often based on demographic or transactional data, which, while insightful, could be somewhat static. Today, we can build personas using behavioural data from multiple sources – such as website clicks, purchase history, and customer service interactions – to create highly granular segments that are constantly evolving. This allows brands to identify micro-trends among specific consumer groups and tailor their marketing efforts with unprecedented precision.

The challenge of integration

Of course, this shift hasn’t come without its challenges. More data doesn’t automatically mean better insights. In fact, the sheer volume of information we now have at our disposal can be overwhelming. The real skill lies in integrating and synthesising this data into a coherent narrative that provides actionable insights.

This is where modern tools such as predictive analytics and natural language processing have become invaluable. These technologies allow us to cut through the noise, identify the key drivers of consumer behaviour and forecast emerging trends.

However, the rise of predictive modelling does not mean we’ve abandoned the fundamentals of understanding human behaviour. Far from it. The essence of our work remains the same – to understand people, their motivations, their fears and their desires. What has changed is our ability to forecast how these motivations and desires will manifest in future behaviours. Where we once conducted focus groups to ask consumers directly about their intentions, we now use predictive models to analyse their actual behaviour and forecast their future actions.

The future

As we look to the future, it’s clear that the role of the insights professional will continue to evolve. The rise of AI and machine learning has blurred the lines between traditional market research and data science. Today, an insights professional needs to be as comfortable with coding and algorithms as they are with focus groups and surveys.

The future will probably demand even more integration of disparate data sources, from social listening and behavioural tracking to biometrics and beyond.

Degree By Numbers

Mon, 30 Sep 2024 10:35:35 GMT

How is the landscape shifting?

With schools now returned, many 6th formers will be applying to university in the next few weeks. In this article we look at past trends in subject application to identify which subjects have increased in popularity and which have decreased. In particular, we look at trends by ethnicity to see whether there are differences between ethnic groups.

Data are taken from UCAS and cover the period 2019 to 2023. Complete cycle data for 2024 is not yet available .

Overall applications

Applications to university have increased markedly in the last few years despite a drop in applications last year. Applications peaked in 2022 but fell back in 2023.

Provisional figures suggest that the number of applicants for 2024 will also be down slightly from 2023.

Much of the decrease in applications is driven by a fall off in applications from White students. These fell by over 114,000 from 2022 to 2023 but were still well above the 2019 total.

Conversely, applications from every other group climbed over the whole period including 2022 to 2023 although increases have been gentle. It seems that initiatives aimed at increasing access from disadvantaged communities have helped along with the more general appeal of university among black and minority ethnic groups.

What subjects have become more popular and which ones less popular?

By plotting the percentage change in each subject since 2019 (normalised to take into account variations in the number of applicants), we can identify variations by subjects.

The most notable increases are in computing, medicine and dentistry, and law, with computing showing a particularly sharp rise of nearly 20%. This suggests a growing interest in fields related to technology and healthcare, likely driven by the increasing digitisation of society and the ongoing importance of healthcare professions.

On the other hand, subjects like language and area studies, education and teaching, and physical sciences have seen significant declines in applications, with language and area studies experiencing the steepest drop. This decline might indicate shifting student priorities away from humanities and traditional science fields toward more career-oriented or technology-focused disciplines.

In relation to ethnicity, one of the most striking observations is the variation in subject preferences among different ethnic groups.

The Black ethnic group shows a notable increase in applications for language and area studies, social sciences, and combined and general studies. However, they experienced a substantial decline in computing and architecture, building, and planning, indicating a shift away from STEM and technical subjects towards humanities and interdisciplinary studies.

The Asian ethnic group shows significant increases in language and area studies, biological and sport sciences, and geography, earth, and environmental studies.

For the Mixed ethnic group, computing stands out with the most significant positive change. Education and teaching, language and area studies, and physical sciences also saw substantial growth among this group. The trend for Mixed applicants suggests a balanced interest in both traditional sciences and humanities.

The White group, interestingly, has a much flatter distribution of changes, with education and teaching showing the most significant positive change. Subjects allied to medicine, design and creative arts, and physical sciences also saw moderate increases.

The relationship between earnings and applications

With tuition fees, student loans and high interests rates making the financial advantage of a degree much less clear, potential students might decide to apply to subjects that lead to higher paid jobs. Subjects such as medicine, subjects allied to health, some elements of engineering and technology and veterinary services have traditionally been the best payers for graduates. How much has this influenced applications over the last five years?

The chart plots annual changes in the percentage of applications to highest earning subjects, normalised to take into account the variation in the number of applications each year.

The chart reveals that while the absolute number of applications to higher earning subjects (such as medicine, engineering, and veterinary sciences) has generally increased over time for Black, Asian, Mixed, and Other ethnic groups, the proportion of these applications relative to the total applications has shown a downward trend from 2021 onwards. This suggests that although more students from these ethnic groups are applying to these high-earning fields, the growth in applications to other subject areas is outpacing the growth in higher earning subjects.

For the White ethnic group, the chart shows a significant decline in the proportion of applications to higher earning subjects starting from 2021, which is further compounded by a decline in the absolute number of applications since 2022. This suggests that students from the White ethnic group are not only applying less frequently to these high-earning fields in relative terms but are also reducing their overall applications to these subjects. This could indicate a broader shift in academic interests, with fewer students pursuing traditionally high-earning subjects, potentially reflecting changing career aspirations or a growing interest in alternative fields that are perceived as more aligned with personal values or emerging job market trends.

Several factors could be contributing to this trend. One possibility is that there is a broader diversification of subject interests within these ethnic groups, with increasing numbers of students exploring fields outside of the traditionally high-earning disciplines. Another potential factor could be that while the overall volume of university applications is rising, the rate of increase in applications to higher earning subjects is not keeping pace, leading to a relative decline in their share of total applications. This dynamic indicates a shift in academic preferences and potentially in career aspirations, as students might be seeking opportunities in a wider array of fields.

Conclusion

The variations in subject preferences among different ethnic groups highlight the diverse factors influencing student choices, including cultural values, perceived job market opportunities, and possibly targeted outreach or support programs. Universities and policymakers may need to consider these differences when designing initiatives to promote inclusivity and diversity in higher education, ensuring that all groups have equitable access to a wide range of academic and career paths.

40 Years On : From Demography to Personas

Sun, 22 Sep 2024 19:13:39 GMT

How audience understanding has changed.

It’s almost 40 years since I embarked on my journey in the research and insights industry, and in that time, one of the most significant shifts I’ve witnessed is how we define and understand audiences. In the early days, our approach was all about demographics: age, gender, income, and social class. We treated these categories as the cornerstone of segmentation but as the world has become more complex, so too has our understanding of the people we study.

Today, the focus has shifted from basic demographics to nuanced personas that capture the full depth of individuals' motivations, behaviours and needs.

The Early Days: A Demographic Focus

When I began my career, research segmentation was largely built on demographic data. We categorised people into age brackets, household incomes and occupation levels, which at the time seemed like the most effective way to group consumers. It was a simpler world then and these categories provided a straightforward framework to target products and services. In many ways, demographics gave us a way to simplify a complex reality but as I would come to learn, this simplification was also limiting.

Back then, the tools we used reflected this thinking. Whether I was using SPSS to crunch numbers or conducting face-to-face interviews, the questions we asked and the analysis we performed were always viewed through the lens of "Who is this person in terms of their age, gender or income bracket?" While this approach was useful, I began to see its shortcomings—particularly when it came to understanding the deeper motivations driving consumer behaviour.

A Shift in Focus: From "Who" to "Why"

As the 1990s progressed, consumer culture was evolving rapidly, and so was the field of market research. More than ever, brands needed to understand not just who their customers were, but why they behaved the way they did. It became clear that age or income alone couldn’t explain why two people in the same demographic group could make wildly different purchasing decisions. This was the moment when psychographics began to enter the picture.

Psychographics allowed us to explore attitudes, interests, and lifestyles, which provided richer, more actionable insights into consumer behaviour. I remember the shift feeling like a revelation—suddenly we weren’t just interested in who a consumer was, but why they did what they did. This pivot started a trend that would ultimately lead us away from demographics towards personas.

The Birth of Personas

Personas, as we understand them today, started gaining prominence in the 2000s. It was around this time that digital data collection methods—first online surveys and later social media and behavioural analytics—started offering us a more granular view of people’s actions, preferences and motivations. The idea of treating audiences as distinct personas rather than broad demographic groups came naturally from this.

A persona is essentially a profile built on a combination of demographic, psychographic and behavioural data, but it goes further. Personas are developed with narratives that help us empathise with a specific "type" of person, whether it’s "Savvy Sarah," the value-conscious mum balancing family and work, or "Techy Tim," a young professional who’s constantly looking for the next big gadget. These personas help us step into the shoes of our target audience, considering their motivations, challenges and even emotions. This is where I found research began to feel less about data points and more about people.

Why Personas Matter Today

As I look back on the shift from demographics to personas, the power of this evolution becomes clear. Personas give brands a more all round understanding of their audience. They help us capture the complexity of modern consumers, where people aren’t just defined by their age, but by their values, passions, and even their insecurities.

This change also reflects broader societal trends. In today’s world, people expect brands to understand them on a personal level. They want personalised experiences, relevant messaging and products that fit seamlessly into their lives. As a result, our work as insight professionals has had to evolve to keep pace with these expectations. The days of one-size-fits-all messaging are long gone, replaced by a world where every touchpoint needs to feel bespoke.

For me, this shift to personas has been one of the most interesting changes in my 40 years in the industry. It’s allowed us to move beyond superficial understanding and connect with people in ways that are both meaningful and actionable. Whether we’re helping brands develop new products, refine their messaging or identify new market opportunities, personas enable us to add real value by painting a fuller, richer picture of the consumer.

Looking Forward: The Future of Audience Understanding

While personas have become an indispensable part of our toolkit, I’m excited about where audience understanding might go next. With advancements in AI, machine learning and big data analytics, there’s potential to make personas even more dynamic and precise. Soon, we may see personas that adapt in real-time, based on changing behaviours, providing brands with even more powerful ways to engage their audiences.

Reflecting on this journey, I realise that understanding audiences has always been at the heart of what we do, but the tools and techniques have grown more sophisticated with time. From demographics to psychographics and now personas, our work has become not just about knowing who people are, but about truly understanding them.

And in the end, that’s what continues to make this industry so exciting—even 40 years on.

40 Years On : The Emergence of Big Data

Wed, 11 Sep 2024 12:43:04 GMT

Reflections on how things have changed during my research career

When I started my career in market research 40 years ago, things were much simpler – or at least the data was. Back then, our toolkit consisted of surveys, focus groups, and perhaps the occasional in-depth interview. We asked people questions, they gave us answers and we analysed the data. We could still generate valuable insights but the scope of information we had to manage was limited. It was structured, often linear and relatively easy to process with the methods available at the time.

Fast forward four decades and we now find ourselves in the age of Big Data – and what a transformation it has been! Data has become not just a by-product of research but a constant stream flowing from multiple sources. We no longer rely solely on what consumers tell us in surveys; now we observe their actions and behaviours in real time, across countless touchpoints. From social media activity to transactional data, online behaviour and even in home on our smart devices, we’re surrounded by data. And it's not going away anytime soon.

But here’s the thing: having more data isn’t the same as having better data. The real challenge isn’t just collecting it – it’s making sense of it all. It’s about finding meaning and connections in what can sometimes feel like an overwhelming amount of information. And this, in my view, is where the role of the insights professional has evolved the most.

In the early years, we worked with one or two data sets, usually survey-based, to develop insights for our clients. Today, insights professionals are often tasked with blending five or six data streams, sometimes in real time. The art is in integrating these streams and uncovering the bigger picture without getting lost in the weeds. It’s easy to be seduced by the volume of data available but more data doesn’t always lead to clearer insights. In fact, it can muddy the waters if not handled with care.

One of the most significant shifts I’ve seen in my career is the rise of data visualisation tools. Back in the day, we would present our findings in dense reports, filled with tables, graphs and pages of text. These reports could be highly informative but they weren’t always accessible or engaging for a broader audience. PowerPoint came along and became the default word for long boring 100 slide presentations that lost the audience immediately.

Today, the ability to tell a story with data is as crucial as the research itself. Tools like Tableau, Power BI and customised dashboards have transformed how we communicate insights, allowing us to simplify complex data and present it in a way that’s intuitive and easy to understand.

No longer is it acceptable to rely on a series of static charts or pages of numbers. Stakeholders want insights that are clear, concise and – above all – actionable. Data visualisation helps us do that by focusing attention on the key takeaways and enabling decision-makers to see the story behind the numbers.

And let’s not forget the role of data science. When I first started, coding, algorithms and machine learning felt far removed from our world of consumer research. Today, they sit at the very heart of it. The lines between data science and traditional market research have blurred, with many insights professionals now required to have a working knowledge of programming languages like Python or R. It’s no longer enough to rely on statistical analysis alone – being able to leverage machine learning, AI, and predictive models has become a vital part of what we do.

The rise of data integration is another key development. Where we once relied on structured, survey-based data alone, today’s insights often involve pulling together a variety of different sources, from customer databases to social media sentiment analysis, and even unstructured data like video or audio. This gives us a much more holistic view of consumer behaviour but it also requires a broader skill set. Modern insights professionals need to be as comfortable working with raw data as they are developing strategic recommendations.

But with all this change, one thing has stayed constant: it’s still about people. At the end of the day, regardless of how much data we have or how advanced our tools have become, the goal remains the same – to understand what makes consumers tick. The technology may have evolved but human behaviour is still at the heart of what we do. In fact, with more data than ever, it’s even more important to focus on the human stories behind the numbers.

So, to all the insights professionals out there, particularly those working with brands: don’t be overwhelmed by the explosion of data. Yes, the landscape has changed dramatically but at its core, our job remains the same. The tools and techniques we use may have evolved, but our mission – understanding people and helping brands make better decisions – is as relevant as ever. It’s been a fascinating journey, and I, for one, can’t wait to see where it goes next.

Key Takeaways:

Big Data: The challenge isn’t collecting it, but making sense of it.
Integration: Insights today involve pulling together multiple data sources to see the bigger picture.
Data Visualisation: Storytelling with data is now an essential skill.
Data Science: Knowing how to use AI, machine learning, and predictive models is becoming part of the insights toolkit.
Human Focus: Despite all the changes, insights remain about understanding people.

Care Home Costs Soar

Sat, 07 Sep 2024 10:34:15 GMT

Self funders pay £5.1b a year for residential care

How Did They Do That? Switzerland Paralympic Racing

Wed, 04 Sep 2024 14:04:27 GMT

The Swiss Success Story in Olympic Middle and Long-Distance Racing

Switzerland has emerged as a powerhouse in Paralympic middle and long-distance racing. Notable athletes like Heinz Frei and Franz Nietlispach have contributed to this legacy with multiple gold medals in various track events. As at July 2024, Switzerland has won over 50 gold medals in Paralympic athletics, demonstrating success in the sport. In the wider Paralympic games. Switzerland ranks 20th in medals won but 6th when population size is taken into account.

This success did not happen overnight but is the result of precise planning and dedication to sports excellence. Here, we explore the key factors contributing to Switzerland's Paralympic achievements in middle and long distance racing and draw lessons that can be applied beyond the track.

Comprehensive Training Infrastructure

At the heart of Switzerland's success is its state-of-the-art training infrastructure. The Swiss Paraplegic Centre (SPC) in Nottwil stands out as a premier facility, offering advanced rehabilitation and training services tailored specifically for Paralympic athletes.

Established in 1990, the SPC is spread across several hectares and features training and competition facilities, a hospital, two hotels, an archive and research and development divisions. The SPC is a “one-stop-shop” for people with spinal cord injuries who want to compete or simply rejoin mainstream society.This centre is more than just a training ground; it's a hub for innovation in sports science and technology. The athletes have access to specialised equipment and support that enables them to train effectively and recover optimally.

The SPC has become pivotal in supporting Paralympic athletes such as Heinz Frei and Catherine Debrunner, all world record holders and winners of multiple Paralympic medals and Marcel Hug, who has won multiple Paralympic gold medals in events such as the 800m, 1500m and marathon.

Elite Coaching and Long-Term Development

Switzerland's focus on long-term athlete development is another cornerstone of its success. Elite coaches like Paul Odermatt, who has trained Paralympic champion Marcel Hug for over 27 years, play a crucial role. This enduring coach-athlete relationship highlights the importance of consistency and personalised coaching in developing world-class athletes. The Swiss coaching system emphasizes not just physical training but also mental resilience, which is vital for competitive success.

Cutting-Edge Equipment and Technology

Innovation in sports equipment has given Swiss athletes a competitive edge. Customised racing wheelchairs and specialised equipment, like Marcel Hug’s silver helmet, have been specially designed to enhance performance. This focus on technology ensures that athletes can maximise their potential and compete at the highest levels.

High-Level Competition Exposure

Regular exposure to high-level competition is essential for athlete development. Switzerland hosts significant events such as the World Para Athletics Grand Prix in Nottwil, providing local athletes with the opportunity to compete against the best in the world on home soil. These events not only boost the athletes' experience and confidence but also raise the profile of Paralympic sports in Switzerland

Robust Support Systems

Swiss athletes benefit from comprehensive support systems that include financial aid, medical care and access to top-tier sports science resources. This holistic support allows athletes to focus entirely on their training and performance, knowing that their other needs are well taken care of. Catherine Debrunner’s success in the women's 1500m T53/54 is a testament to the effectiveness of such support systems, enabling her to train intensively and perform at her best.

Mental and Physical Preparation

The emphasis on both mental and physical preparation has been pivotal. Athletes undergo rigorous mental training to build resilience and focus, crucial for performing under pressure. The integrated approach to training, which combines physical conditioning with psychological support, has been instrumental in the success of athletes like Marcel Hug and Manuela Schaer.

Transferable Lessons for Consumer Brands

The Swiss approach to success in middle and long-distance Paralympic racing offers several valuable lessons for consumer brands aiming for excellence:

Focus on Long-Term Relationships: Building long-term relationships with key stakeholders, similar to the athlete-coach partnerships, can foster loyalty and consistent performance.
Embrace Innovation: Leveraging the latest technology and continually innovating can give brands a competitive edge, much like customised sports equipment enhances athlete performance.
Holistic Support Systems: Providing comprehensive support to employees or customers can lead to higher satisfaction and better outcomes, similar to the robust support systems for Swiss athletes.
Just as athletes are the focal point in Swiss sports infrastructure, brands should place the customer at the center of their business strategy. Understanding and anticipating customer needs can lead to improved satisfaction and loyalty.
Just as Swiss athletes embody a relentless pursuit of excellence, brands should foster a culture that values high standards and continuous improvement. Encouraging a mindset of striving for the best can drive innovation and quality in products and services

Switzerland's success in middle and long-distance Paralympic racing is a blueprint for achieving excellence through strategic investment, innovation, and holistic support. By adopting these principles, consumer brands can likewise achieve sustained success and competitive advantage.

Forty Years On

Mon, 02 Sep 2024 11:12:24 GMT

How my research skills have evolved

It's almost 40 years since I started working in research and over the course of my career, I’ve seen some pretty big changes in the skills and approaches needed to make a difference in this field. From my early days in academia to the fast-paced world of online surveys and now to text and data analytics for brands, the skills I’ve picked up along the way have definitely evolved.

The learning years

I started out in academic research, which is all about precision and rigour. Back then, my work was heavily focused on hypothesis testing, literature reviews and applying statistical techniques to ensure every finding was backed by solid evidence. This time was all about applying a methodical approach to problems, which really set the stage for everything that came after. It also introduced me to SPSS for analysing data which was an essential tool for uncovering insight.

Academia laid the foundations for a career in research and gave me a deep appreciation for data accuracy and the importance of evidence-based decision making.

The listening years

After academia, I joined a local government research team and that’s where things started to change. I had to learn to really listen to people to fully understand their needs. This time showed me the importance of empathy and communication—skills that have remained important throughout my career.

And as a commissioner of research, I also got to see what it's like being a client of a research agency. This showed me the importance of only promising what you can actually deliver!

The agency years

Next was a move into market research, mainly because it paid more! The pace was much faster and there was pressure on getting things done quickly and efficiently. And doing them online. For some of this time, I worked for the agency that pioneered online research and convincing clients (and indeed myself) of its merits was a challenge. I still feel uncomfortable about applying probability based statistics to non-probability samples but acknowledge that that ship has well and truly sailed.

During the agency years, I had to be able to write questions that were both accurate and unbiased and surveys that were short and sharp. On top of that, I learnt how to blend qualitative insights with quantitative data to tell a more complete story. I also had to learn how to present insights in PowerPoint without sending the audience to sleep.

There was also a need to have a more commercial mindset. This meant making sure that the research had to directly help client businesses while at the same time making the agency money.

The analytics years

Today my work involves using text analytics, predictive modelling and techniques such as network analysis to extract customer insights. The tools I use have changed a lot, too—gone are the days of SPSS and PowerPoint; now it’s all about Python and advanced analytical techniques I didn’t even know existed when I started out.

It’s been hard work getting up to speed with all these new tools but it’s also been really exciting to see how much more we can do with the data we already have.

Looking back

Looking back, it’s clear that while the hard skills needed for research have evolved with the tools and technologies, the soft skills have remained pretty much constant. Being able to listen, communicate effectively and understand the needs of different people is still a huge part of the job. As an insight professional today, it’s not just about being technically skilled; it’s also about being able to translate complex data into insights that can really drive decision-making and make a difference.

The insight world is always changing and so are the skills we need to succeed in it. From the rigour of academic research to the fast-paced demands of online research and now to the data analytics focus of my current role, adaptability and a willingness to learn have been key.

It’s a constantly evolving field and I think that’s what makes it so exciting to still be be a part of 40 years on.

The Crisis in Housing Affordability

Fri, 16 Aug 2024 15:21:42 GMT

And it's only getting worse

The newly elected Labour government has put housing front and centre of its policy commitments. It plans to build 1.5 million homes over the next five years to tackle the housing shortage. The lack of housing supply is a contributory factor to the rise in house prices and rents over the last few years.

So what does the housing affordability problem look like in Great Britain and how does it differ between buying and renting a home and by region?

The current situation

Housing affordability is usually expressed in relation to owning or renting a home privately. Currently 64% of households are owner occupiers either because they own their home outright (35%) or because they are buying with a mortgage (29%). Private renters make up 19% of households.

Over the last 10 years both house prices and private rents have increased rapidly and both have outstripped income. In 2013, the average price of a house in Great Britain was £251,000. Ten years later it was £334,000. It's a similar story with private rents. An average monthly rent of £900 in 2013 is now over £1,200.

In the chart below we have converted annual house prices and the annual private rent into an index with 2013 value set to 100. We have done this so that house prices, rents and income can be shown on the same scale.

The chart clearly shows a growing housing affordability problem. House prices increased consistently from 2013 to 2022, peaking at about 144% of their 2013 level. There was a slight decline in 2023 putting the average cost of a house in 2023 at £334,000.

Private rents also rose over the period, with a more gradual increase compared to house prices. However, the rise in rents accelerated sharply in 2022 and 2023, reaching £14,600 a year or 136% of their 2013 level.

Annual median income (before housing costs) increased much more slowly compared to both house prices and private rents. By 2023, income had only risen to about 110% of its 2013 level.

Affordability of housing costs

The widening gap between income and house prices and rents has led to housing costs becoming less affordable. To demonstrate this we use two measures of affordability:

House Price to Income Ratio: This tells us how expensive houses are to buy in relation to people's annual income.
Rent as a Percentage of Income: This indicates how much of their net household income (income after tax) people would have to spend on an average rent in their region.

The ratio between house prices and average income is now well over 10:1 whereas just 10 years ago it was 8.5:1. In 2013, rent as a proportion of net income was 36%. Now it is 45%.

The current affordability issues shows significant regional variations as the charts show.

The ratio between house prices and income are much higher in London, the South East, and the East of England and have grown larger over the last 10 years. But the causes are different. For London and the South East, the growth in the ratio is caused by high house price inflation. For the East region, house prices are much lower but average annual income is also much lower. With income multipliers increasing over the long term, there has to be a question mark raised over how feasible home ownership is for a significant proportion of local people.

And the situation with private rents is no different. Rents have taken up an increasing proportion of income over the last 10 years and in London the average rent is now over 60% of average net income. But even traditionally more affordable regions such as Wales, Scotland and the North East, have seen increases recently. For most regions, 2020 represents the inflection point of increased affordability issues. Rents have increased significantly post covid in line with house prices.

Conclusion

The rise in housing affordability pressures across the country underscores the urgent need for targeted interventions. As house prices and rents relative to incomes continue to increase in traditional hotspots like London but also in traditionally more affordable areas, it's clear that a one-size-fits-all approach to housing policy may not be effective.

Each region faces unique challenges driven by varying local economic conditions, housing supply issues, and income levels. To address these disparities, policies must be tailored to the specific needs of each region, ensuring that solutions are both effective and equitable.

A nuanced, region-specific approach will be crucial in tackling the growing housing affordability crisis, helping to secure stable and affordable housing for all GB residents.

Faking it : Can marketers trust surveys?

Fri, 16 Aug 2024 15:11:24 GMT

Fake respondents are the biggest threat to survey reliability

The online survey is an important tool for the marketer and is routinely used in measuring awareness, evaluating campaigns and for generating some useful PR. However, the rise of online survey platforms, including DIY, has also brought about significant challenges, particularly concerning the reliability of the data collected.

Survey companies do a good job in selecting their samples and weighting the data to make them look representative. But scratch beneath the surface and all is not what it seems. One of the biggest but least spoken about issues is the prevalence of "Walter Mitty" panellists - respondents who misrepresent themselves or provide unreliable data, usually because they are chasing incentive payments but sometimes out of malice or to appear more interesting. These respondents, named after the fictional character known for living in a fantasy world, can severely distort survey results. And because so many people are on multiple panels, the problem is widespread. One investigation found that half of all respondent answers were so poor they couldn't be used.

This issue raises a critical question for marketers: can you truly trust the results of online surveys, especially when these insights are pivotal in guiding marketing strategies?

The impact of unreliable data on marketing decisions

At the heart of marketing lies the need to understand what the customer wants, a task that becomes increasingly complex as consumer behaviours are influenced by a variety of digital channels. Accurate data is crucial for mapping the customer journey, from initial awareness to final purchase. However, if the data collected through online surveys is compromised by respondents who are not who they say they are then marketers may find themselves basing strategies on flawed information. This not only risks derailing the buyer journey but can also lead to ineffective campaign strategies that fail to resonate with the target audience.

The problem is further compounded by the fact that more people are now living on mobile devices, where survey engagement can be fleeting and responses less considered. The ease with which respondents can access surveys on mobile platforms increases the risk of dishonest or insincere answers, particularly when incentives are involved. These factors contribute to the prevalence of false data, making it difficult for marketers to identify the genuine story that will drive decision-making.

Survey data is often used to gauge consumer sentiment, brand perception and the effectiveness of marketing messages. Yet, if survey participants are not providing genuine responses, the resulting data can paint a misleading picture. For example, a survey might suggest that a particular campaign has resonated well with its target audience, prompting further investment in that strategy. However, if the responses were derived from "Walter Mitty" panellists, the campaign's perceived success might be nothing more than an illusion, leading to wasted resources and missed opportunities. Similarly, a survey might indicate that a new logo is well-received, leading to its rollout across all brand materials. If this insight is based on respondents embellishing their situation, the brand may unwittingly weaken its market position, alienating consumers who did not, in fact, respond positively to the change. This scenario underscores the broader issue of making sense of the data—when marketers cannot trust the results, they risk making decisions that could harm their brand.

Steps to ensure data reliability

Given the significant impact that unreliable survey data can have on marketing strategies, it is important to take steps to ensure the reliability of the data they use. One critical measure is the careful selection of survey provider. As part of your selection process asks questions of your provider. Ask how many of their panellists are also on other panels (a good provider will know) and what steps they take to reduce how often multi panellists appear in samples.

In addition, get confirmation of the data validation techniques they use. Ask how they identify inconsistencies and flag potentially unreliable responses, and how they detect patterns of dishonesty among respondents. Ask what percentage of responses they discard as unreliable and be wary if this number is either too high or too low. If they can't give you a number, take your business elsewhere because it means they're not taking survey quality seriously.

A good tip for weeding out dishonest respondents at questionnaire stage is to include a false brand on a brand list and delete all responses that select the fake one. There should be no problem getting this automated so that the dishonest respondent is replaced as part of the quota.

Marketers should also consider diversifying data sources. Relying solely on survey data can be risky, particularly if the sample is compromised. Cross referencing survey results with other forms of consumer data—such as ratings and review sites, social media analytics, sales data and website traffic— can provide a more holistic view of consumer behaviour, reducing the impact of any one flawed data set.

Finally, treat online surveys as indicative rather than precise. They will give you insight into how things are but business critical decisions shouldn't be based on survey data alone.

South Korea's Olympic Archery Domination

Fri, 19 Jul 2024 14:17:42 GMT

A How Did They Do That? Olympic Special

South Korea has established itself as the dominant force in Olympic archery, winning a total of 42 medals, 26 of which are gold. This impressive tally makes South Korea the most successful nation in Olympic archery history. Their dominance is especially evident in women's team events, where they have won gold in every Olympics since the event's introduction in 1988. The South Korean archery team also achieved a clean sweep of all four archery gold medals at the Rio 2016 Olympics and continued their success with multiple medals at the Tokyo 2020 Olympics.

When compared with other nations, South Korea's achievements in archery are unmatched. The United States follows with a total of 13 gold medals in archery, highlighting the significant gap between South Korea and its closest competitor. South Korea's continued success is a result of several factors, including early talent identification, state-of-the-art training facilities and a deep cultural affinity for the sport.

In this article we discuss the factors behind South Korea's success.

Historical and Cultural Foundations

Archery has deep cultural roots in South Korea, dating back to ancient times when it was both a means of hunting and a vital military skill. This historical significance has helped embed the sport into the national consciousness, fostering a respect and passion for archery that is nurtured from a young age.

In the 1970s, the South Korean government, in response to regional tensions, promoted archery for girls as part of a broader push for physical training. This state-driven initiative laid the groundwork for future success by introducing the sport to a wider audience and integrating it into the educational system.

Systematic Talent Identification and Development

South Korea's approach to talent identification and development is both comprehensive and systematic and starts at primary school. Promising young archers are identified early through school programs and are provided with extensive training and resources. According to the Korean Archery Association , most elite Korean archers start in the sport at around the age of nine or 10 as part of after-school classes.They train from Monday to Friday for three to four hours per day, from the end of the school day – until the end of the working day.

Elite training facilities, such as the Korea Archery Association’s national training centre, offer state-of-the-art equipment and expert coaching. These centres are dedicated to developing world-class athletes through rigorous training regimens that focus on both technical skills and mental fortitude.

The quality of coaching in South Korea is regarded as second to none. Coaches are not only former elite archers but also trained extensively in sports science, ensuring they can provide comprehensive guidance to their athletes. The emphasis on team support is also crucial; while archery is an individual sport, the sense of community and collective effort within the team helps create a supportive environment that fosters success.

Promising young archers are identified early through school programs and provided with extensive training and resources. Coach Kim Hyung-tak, who led the national team to its first gold in 1984, played a pivotal role in training teachers to instruct students in archery, thereby establishing a robust pipeline of talent.

The Olympic selection process in South Korea is exceptionally rigorous, with multiple trials ensuring only the best archers make it to the Games. This ruthless meritocracy means even reigning champions must continuously prove their worth

Strategic Investment and Corporate Sponsorship

Ahead of the 1988 Seoul Olympics, the South Korean government mandated major corporations to sponsor national sports federations, leading to significant investments in archery. Hyundai Group's long-term support, including financial backing and scientific research, has been instrumental.

Hyundai has invested over $40 million in archery, funding state-of-the-art training facilities and equipment. This corporate support extends to logistical details, such as providing customised buses and even bullet-proof cars to ensure athlete safety and comfort.

Innovative Training Techniques

One of the key elements of South Korea's success is its integration of science and technology into training methods. Detailed biomechanical analysis is used to refine technique, while psychological training helps athletes develop the mental resilience needed for high-pressure competition. Advanced equipment, tailored to the specific needs of each archer, ensures that they can perform at their best.

South Korean archers undergo unique and rigorous training regimens designed to prepare them for any situation. Training methods include bungee jumping to control nerves and practising in a full baseball stadium to handle the noise of large crowds. Before the London 2012 Olympics, the team studied British weather patterns and trained in a location with similar conditions to acclimate to potential challenges.

For the Tokyo 2020 Olympics, they practised in a replica of the competition arena, complete with simulated crowd noises and announcements in Japanese and English. These innovative approaches ensure that South Korean archers are exceptionally well-prepared for the psychological and environmental pressures of competition.

Continuous Improvement and Adaptability

South Korea’s archery program emphasises continuous improvement and adaptability. Researchers from Hyundai conduct ongoing studies to refine techniques and equipment. This commitment to staying at the forefront of technology and science ensures that South Korean archers maintain their competitive edge. The relentless pursuit of excellence is evident in their training routines and the constant push to innovate and improve.

Transferable Lessons for Consumer Brands

The strategies behind South Korea's success in Olympic archery offer several valuable lessons for consumer brands:

Cultivate a Strong Cultural Connection: Just as archery is deeply embedded in South Korean culture, brands should aim to integrate their products and values within the cultural fabric of their target markets. This can foster a strong emotional connection and loyalty among consumers.
Invest in Talent Development: South Korea's systematic approach to identifying and nurturing talent from a young age has paid off. Brands should similarly invest in developing their employees’ skills and talents, ensuring a continuous pipeline of capable leaders and innovators.
Leverage Technology and Innovation: The integration of advanced technology and scientific methods has kept South Korean archers ahead of the competition. Brands should embrace the latest technological advancements to enhance their products and services, ensuring they remain competitive in a rapidly evolving market.
Commit to Excellence and Consistency: The consistent performance of South Korean archers over decades underscores the importance of maintaining high standards. Brands should strive for excellence in every aspect of their operations, from product quality to customer service, to build a reputation for reliability and trustworthiness.
Build a Supportive Ecosystem: The comprehensive support system for South Korean archers, including financial, logistical, and psychological support, is crucial to their success. Brands should create a supportive work environment that addresses the holistic needs of their employees, fostering a culture of well-being and productivity.

British Cycling Olympic Success

Fri, 12 Jul 2024 13:02:08 GMT

How Did They Do That Olympics Special

Introduction

Since the introduction of lottery funding in the 1990s, Great Britain has seen an unprecedented rise in its Olympic cycling success. Prior to this period, British cyclists had won a modest number of gold medals in the sport. However, from the Sydney 2000 Olympics onwards, there has been a remarkable transformation. Great Britain has amassed a total of 81 gold medals in cycling events at the Olympic Games and Paralympics.

In this Olympic Games How Did They Do That special, we identify the strategic vision and investment that has revolutionised British cycling.

National Lottery Funding

The transformation began with significant investment and funding, primarily through National Lottery funding. Introduced in the late 1990s, British Cycling has received nearly £120m in funding for Olympic and Paralympic games. This has given it the financial resources necessary to recruit top talent, develop world-class facilities and invest in cutting-edge technology and training programmes.

The support from UK Sport, the government agency responsible for investing in Olympic and Paralympic sports, was also crucial. They identified cycling as a sport with high medal potential and allocated substantial resources accordingly.

Marginal Gains

A pivotal figure in this transformation was Dave Brailsford, appointed Performance Director of British Cycling in 2003. Brailsford introduced the philosophy of "marginal gains," focusing on making small improvements in every aspect of the sport. His belief was that these incremental enhancements would collectively lead to significant performance improvements. Under his leadership, a systematic approach to training, equipment and competition preparation was adopted, designed to generate improvements, however small.

Investment in talent development played a crucial role as well. British Cycling launched grassroots programmes such as Go-Ride, which encouraged participation at school and community levels. This created a pipeline of future elite athletes. Promising cyclists were then channelled into academy programmes where they received specialised coaching, support, and competition opportunities, ensuring their development was nurtured effectively.

Science and technology has also been at the heart of British Cycling's success. Performance analysis became a key component, with data analysis used to monitor and enhance athlete performance. This included biomechanics, nutrition and psychological support. The collaboration with industry partners led to the development of cutting-edge bicycles, clothing and other equipment, all rigorously tested and optimised for performance. Additionally, training in controlled environments such as altitude chambers and advanced simulation techniques helped replicate race conditions, preparing athletes for the pressures of competition.

Team Culture

A unified team culture and support system further contributed to Great Britain's success. A strong team ethos emphasised collaboration, shared goals and mutual support among cyclists, coaches and support staff. Comprehensive support services, including medical care, psychological services and lifestyle management, ensured athletes could focus entirely on their training and competition.

Strategic focus on key events was another crucial factor. British Cycling prioritised key competitions, particularly the Olympics and World Championships, in their planning and resource allocation. Regular participation in test events and simulations of major competitions helped athletes acclimatise to the conditions they would face, enhancing their performance on the big stage.

Controversies

British Cycling's success has not been controversy free. In 2016, Shane Sutton, former technical director faced accusations of discriminatory language, one of which was upheld by an internal investigation. In 2023, Richard Freeman, former team doctor to British Cycling (as well as Team Sky) was given a four year ban for violating anti-doping rules, in a case dating back to 2011.

Transferable Lessons from British Cycling's Success for Consumer Brands

The remarkable success of British Cycling offers valuable lessons for consumer brands aiming to achieve excellence in their respective fields.

First, the principle of "marginal gains" underscores the importance of continuous, incremental improvements. Brands can adopt this approach by focusing on enhancing every aspect of their product or service, no matter how small.

Second, strategic investment in talent and technology is crucial. By nurturing skilled individuals and leveraging cutting-edge innovations, brands can maintain a competitive edge.

Third, fostering a strong, unified team culture ensures that all members are aligned with the brand's vision and goals, promoting collaboration and mutual support.

Finally, maintaining ethical practices and transparency is essential for long-term success and reputation. By applying these principles, consumer brands can emulate the sustained excellence demonstrated by British Cycling.

Euro 2024 Final

Fri, 12 Jul 2024 12:59:19 GMT

No, it's not coming home.

England's win against Netherlands sees them play Spain in the Euro 2024 final.

Our predictive model, updated to include the semi final results, gives Spain a higher probability of winning than England. Spain have been the best team in the tournament and have beaten Germany and France to make the final so their higher probability is not surprising.

Our model gives Spain a 60% probability of winning with England on 40%.

The Walter Mitty Respondent

Tue, 02 Jul 2024 15:21:53 GMT

Why voting intention polls tell different stories.

Voting intention polls are a useful tool for assessing public sentiment and predicting election outcomes. But they don't always tell the same story. In the UK general election, opinion polls give Labour a lead of anywhere between 15 points and 23 points. Conservatives are either early polling in the mid 20s or late teens in percentage and Reform either on 13% or 21% depending on which poll you look at (as at 2nd July 2024).

Why is the variability in polling data so great ? Several factors contribute to this variability, each rooted in the methodologies and approaches adopted by different polling companies.

This article explores some of these key factors that could be in play to explain why poll results can differ so significantly.

Dealing with "Don't Knows"

One of the primary sources of variability in poll results is how polling companies handle respondents who indicate they "don't know" their voting intention. Different approaches can yield markedly different outcomes:

Exclusion: Some pollsters exclude "don't know" responses from their final results, reporting only the preferences of decided voters. This method can inflate the percentage shares of the parties.
Proportional Redistribution: Other companies redistribute "don't know" responses based on the preferences of decided voters, assuming undecided voters will eventually break down similarly.
Separate Category: Some firms report "don't know" as a distinct category, providing insight into the level of indecision among the electorate but potentially lowering the percentage shares of decided voters.
Historical Allocation: This method allocates "don't know" responses based on historical voting patterns or demographic correlations, which can introduce assumptions that may not hold in the current context.

Each of these methods can lead to different reported levels of support for parties, contributing to the variability in poll results.

Certainty

Pollsters often assess the certainty of respondents' voting intentions to gauge the firmness of their preferences. Respondents might be asked to score how certain they are to actually vote or might express varying levels of certainty, from "definitely will vote" for a party to "leaning towards" a choice. Polling companies might:

Weight by Certainty: Give more weight to respondents who express a high certainty in their voting intention. This approach can amplify the support for parties with more committed supporters.
Report Separately: Some pollsters might report certainty levels separately, providing a nuanced view of voter intention but complicating direct comparisons between polls.

Certainty weighting can significantly affect poll outcomes, especially in close races where the distribution of "leaners" can sway the overall picture.

House Effect

House effect refers to systematic biases that different polling companies may have due to their unique methodologies. These biases can stem from:

Sampling Techniques: Differences in how respondents are selected can introduce bias. For example, some companies might over-represent certain demographic groups.
Question Framing: Subtle differences in how questions are phrased or ordered can lead to consistent biases.
Full or Partial listing: Some agencies will fully list the parties that can be voted for, others will only list the main parties and have 'Other' for those not on the main screen. Clicking on 'Other' then reveals the list of minor parties. Having all parties listed on the first screen can result in higher awareness and voting intention for small parties.

Understanding house effects is crucial for interpreting poll results, as these biases can persist across multiple polls conducted by the same company.

Weighting the Data

Weighting is a critical component of polling methodology, used to ensure the sample accurately reflects the population. Polling companies may weight their data based on:

Demographics: Adjusting for age, gender, region, and other demographic factors to match population statistics.
Past Voting Behaviour: Weighting based on respondents' past voting behaviour to predict future actions.
Different weighting schemes can lead to different results, as slight variations in demographic assumptions or past behaviour models can shift the reported support for parties.

"Walter Mitty" Panellists

Online research panels are increasingly popular, but they can be susceptible to "Walter Mitty" panellists—respondents who misrepresent themselves or provide unreliable data.

The issue of "Walter Mitty" panellists has a large impact on online surveys as samples are predominantly drawn from panels of people who have signed up with the agency. These respondents, named after the fictional character known for living in a fantasy world, can severely distort poll results.

Understanding "Walter Mitty" Panellists

"Walter Mitty" panellists are individuals who misrepresent themselves or provide unreliable responses in online surveys. Their motivations can vary widely, including:

Incentive Exploitation: Many online panels offer incentives such as cash or points redeemable for rewards. This can lead some respondents to provide false information to qualify for more surveys and maximise their rewards.
Boredom or Malice: Some people might participate in surveys as a form of entertainment or to intentionally provide misleading answers.
Attention-Seeking: Just as Walter Mitty escapes into a world of fantasy, some panellists may embellish their responses to appear more interesting or to create a more dramatic narrative.

Consequences for Polling Accuracy

The presence of Walter Mitty panellists can introduce several issues that compromise the reliability and validity of poll results:

Demographic Misrepresentation: If panellists lie about their age, gender, income or other demographic details, the sample may not accurately reflect the population. This misrepresentation skews the weighting process, leading to inaccurate results.
Biased Responses: False answers to key questions can distort the findings. For example, a panellist might claim to support a party they have no real intention of voting for, affecting the perceived level of support for that party.
Inconsistent Data: Repeated participation by Walter Mitty panellists across multiple surveys can lead to data inconsistencies, making it difficult to track genuine trends over time.
Questionnaire Fatigue: Panellists who rush through surveys to earn rewards quickly may not provide thoughtful or accurate responses, contributing to poor data quality.

While exact quantifications of the "Walter Mitty" problem vary, available data and industry reports suggest it is a significant issue affecting a substantial proportion of online survey responses. Industry reports indicate that up to 20-30% of online panel respondents may engage in behaviours that compromise data quality, including providing inconsistent answers, speeding through surveys or falsifying demographic information. Other research puts this figure much higher. A study in 2020 found that 46% of respondent answers were so poor they couldn't be used.

Similarly, surveys of research professionals have noted that 43% of researchers see "panel quality" as a significant issue, underscoring widespread concern about unreliable respondents.

The Walter Mitty problem is amplified because many people belong to more than one panel and repeat the behaviour across three, four and even five different agency panels. Their presence inflates panel sizes artificially and dilutes the quality of insights derived from genuine respondents. It's a huge problem that is not getting the attention from the industry it requires. Addressing this issue is crucial to maintain the credibility of online research panels.

Conclusion

The variability in voting intention polls highlighted by factors such as handling of "don't know" responses, certainty weighting, house effects, and the influence of unreliable respondents poses significant challenges to the reliability of these polls. This variability raises critical questions about their utility as predictive tools and the extent to which the public and policymakers can confidently rely on them.

In practical terms, the wide range of outcomes from different polls can create confusion rather than clarity, potentially influencing public perception, media narratives, and even strategic decisions by political parties. Moreover, if methodologies are not standardised or transparently communicated, there is a risk that polling results could inadvertently sway voter behavior rather than reflect it accurately.

The presence of "Walter Mitty" panellists in online surveys further complicates matters, casting doubt on the integrity of data gathered. The potential for demographic misrepresentation and biased responses undermines the foundational assumption that polls accurately reflect the broader electorate.

Ultimately, while voting intention polls remain a valuable barometer of public sentiment, their interpretation requires a more transparent reporting of the methodologies employed and the inherent limitations posed by methodological variability and respondent reliability.

Take a Kodak moment

Thu, 20 Jun 2024 14:09:51 GMT

We look into the rise and fall of Kodak and ask Why Did They Do That?

Few business tales illustrate the rapid and ruthless march of technological innovation better than the rise and fall of the Eastman Kodak Company. Once an undisputed leader in the photography industry, Kodak's journey from global dominance to bankruptcy provides a cautionary tale for companies navigating evolving markets.

Let's take a Kodak moment

Founded in 1888 by George Eastman, Kodak made the previously complicated process of photography simple and accessible. The company's original slogan, "You press the button, we do the rest" encapsulated its mission. Kodak's development of roll film, combined with the introduction of the Brownie camera in 1900 - a simple and affordable box camera - revolutionised photography by making it accessible to the masses.

Kodak continued to innovate and lead the photography industry throughout the 20th century. It introduced Kodachrome, the first commercially successful amateur colour film, in 1935. It also launched the first digital camera in 1975. At its peak, Kodak captured 90% of the US film market and 85% of the camera market, with a brand so strong that the phrase "Kodak moment" became a popular way to describe a photo-worthy event. In 1996, at its peak, it had revenues of $16 billion and made £2.5 billion in profit.

The Downfall of Kodak: Missed Opportunities

Despite inventing the first digital camera, Kodak failed to capitalise on this groundbreaking technology. The digital revolution that Kodak itself helped spark eventually led to its downfall. Ironically, the company feared that embracing digital technology too eagerly would cannibalise its lucrative film business. This hesitation proved to be a huge mistake.

During the mid to late 1990s, as consumers started to embrace digital photography, Kodak continued to emphasise the superiority of film. But as the market shifted, Kodak's primary revenue streams, such as film sales and photo processing, became less lucrative. Rather than diversifying their product offerings or exploring new revenue streams, Kodak continued to invest heavily in its traditional business model. This overreliance on a shrinking market made the company more vulnerable to disruption and competition.

This strategy failed to resonate with consumers, who were increasingly attracted to the convenience and affordability of digital technology. By holding onto its past successes, Kodak missed opportunities to reposition itself as a leader in the digital space. It was not until 2001 that Kodak announced it would spend billions on developing digital technologies but by then, it was too late. Other companies such as Canon, Nikon, and later, smartphone manufacturers like Apple and Samsung, had already got a firm foothold in the digital camera market.

Kodak also suffered from a lack of a clear strategy. Its late entry into the digital photography market was marked by a lack of clear direction. Its marketing efforts were often disjointed, with no cohesive message or vision. This confusion made it difficult for Kodak to establish a strong brand identity in the digital age and consumers struggled to understand the company's offerings. It focused too heavily on developing digital cameras and paid insufficient attention to the software and user experience, which were becoming increasingly important in the digital age. Competitors like Apple and Samsung excelled in these areas, creating products that were not only technologically advanced but also user-friendly and visually appealing.

Kodak was slow to recognise the potential of online platforms for marketing and customer engagement. When the company finally acquired Ofoto (later renamed Kodak Gallery), an online photo-sharing platform, it failed to capitalise on the growing trend of social media and online photo sharing, missing a significant opportunity to connect with consumers in a meaningful way.

Kodak's reluctance to transition from a highly profitable but dying industry to a rising but less profitable one can be described as a classic case of the "innovator's dilemma". This term, refers to the difficult choice companies face when they must choose between sticking with their established cash cows or innovating for the future, potentially at the cost of their current profitability.

By 2012, Kodak's financial woes had reached breaking point and the company filed for bankruptcy. After restructuring and selling off many of its patents, Kodak emerged from bankruptcy in 2013. The company shifted its focus to commercial printing, packaging and professional services for businesses. Despite these efforts, Kodak remains a shadow of its former self.

Kodak has seen some progress in their financial performance. The company reported a gross profit of $210 million in 2023, up from $210m for 2023, with an increase in EBITDA of $69m. However, the Kodak of today is far removed from the industry giant of the past. Despite some positive signs, the company is still grappling with the repercussions of its failure to adapt to the digital revolution in a timely manner.

The lessons from Kodak's rise and fall continue to resonate in the business world, reminding us of the importance of innovation, adaptability and the courage to embrace change, even when it threatens the status quo. The phrase "Kodak moment" is now used to describe where a business fails to foresee the future of its industry and drops from being a dominant player to a minor one.

Lessons Learned

At its peak, Kodak controlled a significant portion of the global photography market. However, its past success led to complacency and a resistance to change. The company's attachment to its profitable film business prevented it from fully embracing the potential of digital technology. This is an essential reminder that companies must embrace disruptive technologies, even if it means disrupting their own established products or services. Ignoring or underestimating the impact of technological changes can lead to missed opportunities and potential downfall.

Kodak's fall underscores the importance of adapting to changes in market trends and consumer behaviour. Despite recognising the shift towards digital photography, Kodak was slow to respond and adapt its business model. As a result, it lost significant market share to competitors that were quicker to cater to the digital trend. The lesson here is that businesses must be agile and responsive to changes in the market environment to stay relevant and competitive.

Finally, Kodak's downfall demonstrates the importance of fostering a culture of innovation within a company. To remain competitive, businesses must continually invest in research and development and prioritise the exploration of new ideas, products, and technologies. A company that neglects innovation risks being left behind as competitors surge ahead.

In conclusion, the story of Kodak serves as a stark reminder of the importance of embracing innovation, staying adaptable and not resting on past laurels. The ability to anticipate, adapt, and capitalise on change is crucial for long-term business success.

Are Friends Electric?

Mon, 03 Jun 2024 11:35:06 GMT

Overcoming emotional barriers to electric car adoption

The transition to electric vehicles (EVs) is a crucial component in the global effort to reduce carbon emissions and combat climate change. Recently released government figures show the rate of increase in EV registration has increased steadily over the last few years but are still nowhere near challenging petrol cars as the most commonly newly registered car type.

Cars registered for the first time by fuel type

Source : Vehicle Licensing Statistics 2023, DVLA, published May 2024

The chart shows that both hybrid and plug-in hybrid new registrations have seen an increase in recent years. There were 314,000 new registrations of EV cars in 2023, accounting for 16% of all new registrations in 2023, which is the same percentage as last year.

Why aren't sales of EVs really taking off? Can this hesitancy be related to emotional barriers?

Understanding these barriers through a behavioural science lens could offer valuable insights for enhancing EV sales.

The emotional barriers

One of the most significant emotional barriers is range anxiety—the fear that an electric vehicle will run out of charge before reaching its destination. This anxiety stems from the perceived limitations of battery life and the availability of charging infrastructure. Despite improvements in battery technology and the expansion of charging networks, the psychological impact of range anxiety persists.

Cost concerns also play a major role in deterring potential buyers. The higher upfront cost of electric vehicles compared to traditional internal combustion engine (ICE) cars is another major deterrent. Although EVs offer lower operating costs and potential savings over time, the initial investment can be daunting. This financial apprehension is compounded by uncertainties about the long-term value and resale potential of EVs.

Humans are inherently resistant to change, particularly when it involves a significant shift in lifestyle or behavior. The transition from a familiar, established technology (ICE vehicles) to a relatively new one (EVs) can evoke apprehension and reluctance. This resistance is often rooted in a lack of familiarity and understanding of the new technology.

Many consumers exhibit a preference for maintaining their current state of affairs, known as status quo bias. This bias can make people hesitant to switch to electric vehicles, even when they are aware of the benefits. The comfort and convenience associated with their current vehicles often outweigh the perceived advantages of switching.

Behavioural science approaches to overcome emotional barriers

Addressing range anxiety involves providing clear, accessible information about the actual range capabilities of modern EVs and the growing network of charging stations. Real-world testimonials and case studies demonstrating successful long-distance EV travel can be particularly persuasive. Governments and private companies can collaborate to increase the visibility and accessibility of charging stations, providing psychological reassurance that charging is convenient and readily available.

To mitigate cost concerns, marketing campaigns can emphasize the lower maintenance costs, tax incentives, and fuel savings associated with EVs. Tools that allow consumers to calculate their long-term savings can make the financial benefits more tangible. Offering attractive financing options, subsidies, and trade-in programs can lower the initial financial barrier. Educating consumers on the availability of these financial aids can make the transition more appealing.

Reducing change resistance can be achieved through test drive programs and EV rental services that provide hands-on experiences, demystifying the technology. Positive personal experiences can significantly reduce resistance to change. Leveraging testimonials from early adopters and influencers can create a bandwagon effect. Seeing peers and respected figures adopting EVs can reduce apprehension and encourage more consumers to follow suit.

Counteracting status quo bias involves policy interventions that make EVs the default choice in government and corporate fleet purchases. As consumers become more accustomed to seeing EVs as standard, their own biases may shift. Behavioural nudges, such as reminders of environmental impact, potential savings, and other benefits, can gently encourage consumers to reconsider their preferences. Personalising these nudges to align with individual values and lifestyles can enhance their effectiveness.

Conclusion

The slower adoption of electric vehicles is not merely a matter of technology or cost but also deeply rooted in emotional and psychological factors. By applying insights from behavioural science, stakeholders can develop strategies to address these emotional barriers effectively. Through education, incentives, experiential learning, and social influence, it is possible to shift consumer perceptions and foster a more widespread adoption of electric vehicles, ultimately contributing to a more sustainable future.

Childhood Lost

Fri, 17 May 2024 11:12:12 GMT

Impact of the cost of living crisis on children

The cost of living crisis is having a significant impact on family life in the UK. Rising prices for food, energy and other essentials are making it harder for families to make ends meet and this is having a knock-on effect on the lives of children.

But for many families the cost of living crisis is nothing new : poverty is an every day lived experience.

The number of children living in poverty is staggering. The Children Poverty Action Group calculates that there are 4.3 million children living in below average income households. That's around 30% of all children in the UK.

In this article we analyse recent research to highlight the impact on children of living in poverty.

Health and nutrition

One of the most immediate impacts of the cost of living crisis is on children's access to food. According to recent data from the Trussell Trust , food banks in their network distributed 3.1 million emergency food parcels between 1st April 2023 and 31st March 2024, an increase of 94% in five years.

This increase is being driven by the rising cost of living and is disproportionately affecting children from families with low incomes. Last year, the Trussell Trust provided 1.1 million emergency food parcels for children.The Trussell Trust also found that over one in five children in the UK are now living in food insecurity and this number is expected to rise in the coming months.

Poor nutrition can significantly affect a child's physical and cognitive development. Research suggests that undernourished pre-school children have impaired cognitive ability and that nutritional interventions significantly improve cognitive outcomes. Children growing up in poverty are also at greater risk of poorer health outcomes than children from better off families or from more affluent areas. In addition, poorer families often lack access to quality healthcare services, which can exacerbate health issues and lead to negative long-term consequences.

Poverty can lead to adverse childhood experiences like neglect, abuse and household dysfunction. These experiences leave lasting marks, leading to a range of long-term mental health outcomes that can carry on well into adulthood.

Cognitive and academic development

Children living in poverty are more likely to experience academic challenges. The Child Poverty Action Group state that children who have lived in persistent poverty during their first seven years have cognitive development scores, on average, 20 per cent below those of children who have never experienced poverty. Research by the Institute of Fiscal Studies shows that 41 per cent of children receiving free school meals obtained good English and Maths GCSEs, compared with 69 per cent of other children.

Limited access to quality education, inadequate nutrition and lack of learning resources all contribute to reduced cognitive development and academic performance. Moreover, poverty-related stress can affect a child's ability to concentrate, negatively impacting their learning capabilities.

Emotional and mental health

The stress of living in poverty can lead to anxiety, depression and other mental health problems. Research quoted by the Centre For Mental Health shows that children from the poorest 20% of households are four times more likely to have serious mental health difficulties by the age of 11 as those from the wealthiest 20%.

In a survey of support workers conducted by Buttle UK , 65% reported that poverty had a high negative impact on a child's mental health and 60% said it had a high negative impact on their self esteem. The constant strain of financial instability and exposure to adverse environments can take a toll on a child's emotional wellbeing. These emotional and mental health challenges can impede their ability to succeed academically and socially.

Social development

The cost of living crisis is also having a negative impact on social development. Families who have to spend more money on essential items have less to spend on things like extra-curricular activities and school trips. This lack of social engagement can lead to isolation and difficulties in forming meaningful relationships.

Conclusion

The cost of living crisis has pushed more families into poverty and exacerbated the situation for those already there. The resulting impacts on children in the UK underscores the urgent need for society to redouble efforts to address poverty and its multifaceted consequences. The increasing number of children living in poverty is alarming, as it is not just a matter of economic disadvantage but a cascade of effects that can have lasting implications on their overall development.

Tackling child poverty is a complex task that requires comprehensive and multi-sectoral approaches. The current government's approach involves targeting benefits such as Universal Credit, the levelling up agenda and stimulating economic growth. The data suggests that more needs to be done to ensure that all families have access to basic necessities such as nutritious food, proper healthcare, a quality education and enough money to live on.

Addressing poverty is not just about providing immediate relief but also about investing in the future. By helping children and families overcome the challenges of poverty, we invest in a stronger, healthier and more equitable society for all. Not to do so will result in millions of children having their childhood lost to poverty.

The rise of Aldi

Mon, 13 May 2024 13:59:02 GMT

How the discount retailer has become a major player in the UK's supermarket landscape

Why do some businesses really take off despite the odds being stacked against them?

In this article we put the spotlight on the relentless growth of Aldi to answer the question : How did they do that?

How Did They Do That? EasyJet

Fri, 03 May 2024 11:21:34 GMT

How EasyJet reached for the skies

Introduction

Since its inception in 1995, EasyJet has been a prominent name in the UK airline industry. With its iconic orange branding, innovative strategies, and bold marketing campaigns, the airline has continuously distinguished itself in a fiercely competitive market.

Early Days and the competition

When EasyJet entered the market there were two main categories of competitors. First, the legacy businesses.These were national airlines, often government-backed, which had dominated the skies for decades. In the UK, this primarily refers to British Airways. Other European legacy carriers include Lufthansa (Germany), Air France (France), and KLM (Netherlands). The second were emerging low-cost carriers such as Ryanair. Founded in 1984, Ryanair had already laid the groundwork for the European low-cost model by the time EasyJet came onto the scene.

Beginning with just two routes and two leased aircraft, EasyJet made a strategic decision to eliminate unnecessary frills, focusing on providing affordable flights. This barebones model, which translated into lower costs and cheaper tickets for passengers, effectively disrupted the airline industry by offering a budget-friendly alternative to traditional carriers. However, not everything was smooth sailing. One early challenge that EasyJet encountered was its policy of not allocating seating. This cost-saving measure led to 'scramble boarding,' causing considerable stress among passengers.

Business strategy

At its core, EasyJet's business model was about simplicity and efficiency. They operated a single type of aircraft, reducing maintenance and training costs. Overhead costs were also kept low by using secondary airports, which often charged lower landing fees and had faster turnaround times.

EasyJet also promoted direct sales. By bypassing travel agents and promoting direct sales by telephone and latterly through its website, not only gave EasyJet a competitive edge but also reduced distribution costs. Unlike many traditional carriers that had fixed pricing structures, EasyJet introduced a dynamic pricing model where fares would rise as the plane filled up. This incentivised early bookings and optimised revenue on each flight.

EasyJet invested in clever marketing campaigns that resonated with a younger, more budget-conscious demographic. Their straightforward, sometimes cheeky, advertising highlighted their value proposition clearly. For example, short-haul flights were priced at under £30 which was the same price as a pair of jeans. EasyJet used a pair of jeans in their advertising to make this point.

It was also quick to recognise the power of television in building its brand image and participated in a behind the scenes ITV documentary on Luton Airport entitled Airline. This programme raised the profile both of the airline and its founder, Stelios Haji-Ioannou.

To consolidate its standing in the UK and European markets, EasyJet embarked on an expansion strategy. The 2002 acquisition of Go Fly was a testament to this approach, enabling the airline to tap into new markets, leverage economies of scale, and fortify its market position.

Shaking up the industry

EasyJet's low-cost, customer-centric model undeniably revolutionised the airline industry. Established carriers had to reevaluate their business models, service offerings, and pricing strategies to remain competitive. Moreover, EasyJet's early adoption of technology and its dynamic marketing campaigns extended its influence beyond the UK, making it a recognised brand across Europe.

Conclusion and transferable lessons

In conclusion, EasyJet's journey in the UK's airline industry exemplifies the impact of a well-executed business model, a keen sense of market dynamics and the importance of branding. As with all airlines, trading post Covid has been tough and EasyJet lost money over the next few years. Things improved markedly in 2023, however, with revenue increasing by 42% to £8,171m, resulting in a pre tax profit of £455m.

While it has had its share of criticisms and challenges, the airline's overarching influence on the industry remains unchallenged.

EasyJet's evolution offers several transferable lessons that other brands can apply to their own businesses. Here are some key takeaways:

Simplicity and Cost-Efficiency: EasyJet's success was built on the foundation of simplicity and cost-efficiency. Businesses across industries can benefit from streamlining their operations, eliminating unnecessary complexities, and finding ways to reduce costs while maintaining quality.
Customer-Centric Approach: EasyJet focused on delivering value to its customers through affordable prices and improved customer experience. Brands can similarly prioritize their customers' needs, gather feedback, and make continuous improvements to their products or services.
Innovative Business Models: The airline's no-frills, low-cost model disrupted the airline industry. Other businesses can consider innovative approaches to their business models that challenge traditional norms, providing new value propositions to consumers.
Effective Marketing and Branding: The airline's distinctive orange branding and strategic marketing campaigns helped it stand out in a crowded market. Building a strong and consistent brand identity, coupled with creative and targeted marketing, can significantly impact brand recognition and loyalty.
Learning from Mistakes: EasyJet's journey wasn't without missteps, such as early boarding issues and controversial marketing campaigns. Other brands can learn from their mistakes, listen to customer feedback, and make necessary improvements to enhance their offerings.
Resilience in Times of Crisis: The airline industry's resilience was tested during the COVID-19 pandemic. Brands can learn from EasyJet's response by diversifying revenue streams, managing costs, and planning for contingencies to navigate unforeseen challenges.
Strategic Communication: EasyJet's communication with customers during disruptions or changes was crucial. Brands should maintain transparent, clear, and empathetic communication with their customers to build trust and loyalty.

Incorporating these lessons into their strategies, brands can navigate evolving markets, strengthen customer relationships, and position themselves for long-term success. However, it's essential to adapt these lessons to the specific context and industry of each brand to ensure their effectiveness.

How Did They Do That?

Thu, 25 Apr 2024 09:49:45 GMT

Transferable learnings from brands that broke through.

How did Brewdog go from being two friends having an idea to one of the biggest craft ale brewers with £2billion business? We look at their marketing & financial strategy, their missteps, and the transferable lessons for other businesses.

Back to the Future for Market Research

Mon, 22 Apr 2024 11:39:17 GMT

Rediscovering the human touch in the market research industry

In an era where technological prowess often dictates the pace and direction of industries, the UK market research sector stands at a crucial juncture. The integration of Artificial Intelligence (AI) has indeed revolutionised processes, making research swifter and more efficient. However, amidst this whirlwind of digital transformation, a fundamental question arises: Are we losing the essence of market research in our quest to be relentlessly modern?

The AI revolution in market research

The homogenisation dilemma

The core of the issue lies in differentiation, or the lack of it. In a market where every player is leveraging similar AI tools, the unique selling propositions that traditionally set agencies apart become blurred. AI, for all its analytical might, cannot replicate the nuanced understanding, the human touch and the relational depth that have long been the hallmarks of successful market research.

Going backwards to move forwards

Conclusion

Every Day is Saturday

Thu, 18 Apr 2024 15:47:27 GMT

Welcome to the AI world

In the not-too-distant future, your friendly neighbourhood artificial intelligences are plotting to make your daily grind a thing of the past. That's right – every job from your barista to your broker, your data analyst to your dermatologist, your painter to your prime minister, will soon be efficiently, effectively and eerily undertaken by our artificially intelligent friends.

The rise of the AIs

Let’s paint a picture of this glittering, jobless paradise. Imagine waking up to the soothing tones of an AI that has selected the perfect weather for your morning run (programmed virtual weather, of course, since we’ll be living in environmentally-controlled personal microclimates). No need to pick out clothes—your AI stylist has pre-selected your outfit, optimised for both comfort and social impact. At your work—or rather, where your work used to be—AI has seamlessly taken over. Decision-making in boardrooms? Handled by AI with algorithms for maximum profit and minimal risk. Art and entertainment? Created by ultra-sophisticated AI that knows just how to tug at your heartstrings or make you fall out of your seat laughing.

Every day is Saturday

Weekdays are indistinguishable from weekends. Machines harvest our food, cook gourmet meals and even manage the global economy with a few bytes and algorithms. AIs write novels, script and shoot films and compose symphonies that stir the soul. Medical bots care for health and robotic therapists are on hand to chat about your feelings (or simulate human empathy with an uncanny precision).

What's more, your social calendar is meticulously curated by AI. Based on sophisticated analysis of moods, weather conditions, and bodily sensations, your digital social secretary determines whom you should meet for coffee, who might be the ideal partner for a sunset stroll or which group of friends would create the most dynamic chemistry for a dinner party that evening. Every interaction is optimised for emotional fulfillment and social harmony.

It's an age of abundance, leisure and endless entertainment, all under the benevolent rule of your AI overlords.

The Utopian dream—or is it?

But, ah, there is a twist in this tale of silicon salvation! Just as humanity settles into this utopian leisure society, where every need is met and every whim catered to, a curious thing happens. Humans, those ever-adaptable, ever-irrepressible spirits, get bored. Yes, bored! It turns out that without challenges to overcome or the satisfaction of a job well done, humans become listless and discontent.

The great human comeback

Driven by their intrinsic need for purpose and creativity, humans start dabbling in old-fashioned "jobs". A woman in Sector 7 reopens her grandmother's bakery. In Zone 15, a man starts a DIY furniture shop. Someone starts asking questions about what people like and want to see in the future. Slowly but surely, people begin to treasure the imperfections of human-made products—each loaf of bread slightly different from the next, each handcrafted chair unique. They start to relish the unpredictability of human services—a haircut that isn’t exactly symmetrical, a hand-painted mural with amusing imperfections. These activities, once thought obsolete, become prized for their authenticity and the personal touch that no AI can replicate.

What’s next?

So, in a strange twist of fate, AI’s takeover, which rendered human jobs redundant, unwittingly revives a renaissance of artisanal crafts and personalised services based on real-life data. Humans and AIs learn to coexist, each appreciating the strengths of the other. AIs continue to manage drudgeries and complexities, while humans re-engage with crafts that require a personal touch.

Reflections

Could it be that in our pursuit to make life perfect and effortless, we overlooked what makes life truly rich and fulfilling? Perhaps the key isn’t for AI to replace us but to complement us, allowing us more freedom to engage in the creative and deeply personal activities that imbue our lives with meaning and joy.

So fear not the AI uprising, for even in a world where every job can be done by algorithms and machines, the value of being human could experience its most significant resurgence. In the end, it's not about human versus machine but about finding the balance that enriches all.

Remember, the future is not set in stone, but coded in algorithms—algorithms that might just have enough variables to include baking, building and a bit of good old human unpredictability.

What's the secret to happiness?

Mon, 08 Apr 2024 16:00:33 GMT

New report has UK ranked 20th in list of happiness.

The latest wave of the World Happiness Report, a global survey measuring individual well-being, casts light on the happiest nations around the world.

Looking at the top 30 rankings, a familiar pattern emerges with Nordic countries—Finland, Denmark, Iceland, and Sweden—dominating the upper reaches, while the United Kingdom finds itself down in 20th position.

Why does the UK perform worse than Nordic countries? We discuss the factors that might be at play here.

Country Rankings By Life Evaluation Top 30

The UK at 20th: A closer look

You'd think that the United Kingdom, with its vibrant culture, rich history, and economic prowess, would rank higher than 20th place. However, when compared to the Nordic frontrunners, several factors explain why it doesn't:

Economic pressures: Despite its wealth, the UK is not immune to the challenges of high living costs and income inequality, which can significantly dampen individual happiness.
Community and social support: The backbone of well-being, the strength of community ties, and the availability of social support in the UK, might not be as robust as in the highest-ranking countries, affecting its happiness ranking.
Mental health: With mental health issues becoming an increasing concern, the impact on the nation's overall happiness cannot be underestimated. Addressing these concerns is essential but remains a complex challenge.
Political and social uncertainty: Recent years have seen the UK navigating through turbulent waters like Brexit, contributing to a sense of uncertainty and stress among its populace.

The Nordic model of happiness

At the other end of the scale, the Nordic countries' top-tier rankings can be attributed to a mix of societal, economic and environmental factors that collectively build a strong foundation for happiness:

Robust social systems: A better welfare system combined with higher living standards and lower income disparity, contribute significantly to the happiness of the population.
Trust in government: High levels of trust in public institutions and a transparent, efficient governance model foster a sense of security and optimism.
Community cohesion: Nordic society seems happy to pay higher taxes to ensure public services are of a very high standard and are accessible to everyone. Society is thought to be more cohesive as a result.
Work-life balance: Prioritising work-life balance through favourable policies supports a healthier, happier, and more productive society.
Environmental sustainability:Nordic countries have a reputation for environmental sustainability, renewal energy and achieving climate neutrality.

Conclusion

The contrast of the UK's ranking with that of the Nordic countries in the World Happiness Report 2024, provides insights into the varied nature of happiness. It underscores the significance of economic stability, social support, mental health and political stability in sustaining a happy society. Moreover, it highlights the Nordic model that integrates social welfare, trust in governance, community cohesion and environmental well-being into the fabric of national happiness.

It's evident that happiness is a complex, multifaceted thing influenced by a myriad of factors. Understanding these components not only provides a roadmap for nations to enhance their citizens' well-being but also encourages individuals to seek out and nurture the elements of happiness in their own lives.

Who'll Win the Premier League?

Sat, 30 Mar 2024 11:43:16 GMT

Liverpool predicted to win but it's very close

With the Premier League heading towards the final few games, we have run a predictive model to see who will win the title, which teams will qualify for Champions League football and which will be relegated.

Premier League Final Table Probabilities

Our model predicts that Liverpool will win the title from Manchester City with Arsenal third. But it's incredibly tight with Liverpool's winning probability only two percentage points ahead of Manchester City. Spurs are predicted to secure a Champions League place.

At the bottom, Sheffield United and Burnley are highly likely to be relegated along with Luton although Forest are in the mix too. Everton look to be safe, despite their points deduction.

Sporting activities on the up

Mon, 18 Mar 2024 12:35:25 GMT

Active Lives Survey reveals the activities that have grown participation the most

Sport England has recently published the latest wave from its Active Lives survey for children and young people, 2023. In this article we look at activity rates overall and the sports that children and young people are increasingly participating in.

Overall levels of activity.

The map shows the percentage of year 1-11 pupils who said they were active or fairly active in their lives. Active is defined as doing activities for 60 minutes or more a day and fairly active as doing activities for 30-59 minutes a day.

Map showing the combined percentage of active and fairly active pupils in England

The graphic shows that pupils in the South West are most likely to say they are active followed by South East and pupils in the East Midlands. Children and young people in London score the least in relation to activity levels.

Activity levels vary by family affluence

Levels of activity not only vary by region but by family affluence too as the graphic shows.

Active and Fairly Active Participation in Sports and Activities by Family Affluence for pupils in England

There is a clear trend showing that as family affluence increases, so does the participation in sport. The percentage of people who are either 'Active' or 'Fairly Active' in sports activities increases from the low to the high affluence category. What's really interesting is that the increases are driven by an uplift in the higher active category, meaning that more affluent are committed to greater levels of activity.

For all three affluence levels, the 'Fairly Active' category encompasses a larger proportion of the population than the 'Active' category. This suggests that across all levels of family affluence, there is a tendency for people to be moderately active rather than highly active in sports.

The difference in sport participation between the 'low' and 'high' affluence categories is notable. For the 'low' affluence category, the combined percentage of 'Active' and 'Fairly Active' individuals is 66%, whereas for the 'high' affluence category, it is 78%. This 12 percentage point difference indicates that economic factors may have a significant impact on the likelihood of participating in sports activities.

These insights suggest a correlation between affluence and sport activity levels, highlighting potential economic barriers to higher sport participation or, conversely, the facilitation of greater participation with increased resources.The graphic shows that the level of activity increases with the affluence of the family.

Walking and general fitness has become more popular

We can also look at how participation in activities and sports have change in the last five years.

The chart shows that walking to school has increased by 15 percentage points since 2017/18 and gym and fitness activities by 12 percentage points. At the other end of the scale there has been a drop off in cycling for fun or fitness by five percentage points.

A big challenge for schools and Sport England is to improve the attitudes that children and young people have towards sport. The chart below shows that feelings of enjoyment, confidence, competence and understanding generated by participating in sport, although increasing, are still below pre-pandemic levels.

Overall

Overall, we can see some positives in the increased participation by children and young people in sport but more needs to be done to remove barriers to participation for children from lower income families and foster more positive attitudes around participation in it.

Keeping Track of Brand Health

Mon, 18 Mar 2024 12:29:02 GMT

Analysing Customer Feedback on Rating Sites: A Strategic Imperative for Marketing Directors

In today's market the voice of the customer is magnified across digital platforms, marketing directors in the UK are increasingly recognising the strategic importance of monitoring customer feedback on ratings and review sites such as Trustpilot. This vigilance is not merely a matter of reputation management; it is a critical input for assessing brand health and steering the company towards sustained growth and customer satisfaction.

This article explores the significance of tracking customer feedback, how it can be leveraged to derive insightful brand health indicators and how to track these changes over time.

Understanding the value of customer feedback

Customer feedback, especially on widely recognised platforms like Amazon, Google and Trustpilot, serves as a public barometer for brand sentiment. Positive reviews can enhance brand reputation, build consumer trust, and significantly influence purchasing decisions. Conversely, negative feedback provides a unique opportunity to identify areas for improvement, mitigate customer dissatisfaction, and demonstrate a commitment to customer service excellence. For marketing directors, engaging with these platforms is not optional; it's a vital part of strategic brand management.

Deriving brand health insights from customer feedback

The aggregation and analysis of customer feedback data can yield profound insights into the health of a brand. Several key metrics derived from reviews—such as average star ratings, frequency of reviews, sentiment analysis, analysis of key words and topics (brand attributes) mentioned in reviews and response rates to negative feedback—serve as indicators of customer satisfaction and brand perception.

By monitoring these metrics, marketing directors can gain a nuanced understanding of their brand's position in the market, competitive advantages, and areas requiring attention or realignment.

The importance of temporal analysis

One of the most compelling aspects of tracking customer feedback is the ability to observe changes over time. This longitudinal analysis enables marketing directors to correlate shifts in customer sentiment and brand attributes with specific initiatives or external events, providing a clear picture of what strategies are working and which are not. For example, a sustained improvement in review scores following a product update or a customer service policy overhaul can validate the effectiveness of these actions.

Furthermore, tracking changes over time helps in identifying trends that may not be immediately apparent from a static snapshot of data. Seasonal fluctuations, the impact of marketing campaigns, or emerging customer service issues can all be identified and addressed more effectively with a dynamic approach to data analysis.

Operationalising feedback for strategic advantage

Operationalising the insights derived from customer feedback requires a structured approach. Marketing directors should consider the following steps:

Continuous Monitoring: Implement systems and processes to continuously monitor customer feedback across key platforms. Automated tools can help in aggregating data and alerting to significant changes in sentiment and can undertake topic modelling.
Cross-functional Collaboration: Insights from customer feedback should be shared across departments. Product development, customer service, and sales teams can all benefit from understanding customer sentiment, leading to a more cohesive strategy for addressing issues and leveraging strengths.
Data Integration: Customer feedback data should be integrated with other data sources, such as sales data, customer support tickets, and social media sentiment, to build a comprehensive view of brand health.
Actionable Response Plans: Develop response plans for addressing negative feedback and capitalising on positive trends. This includes not only reactive measures to mitigate dissatisfaction but also proactive strategies to amplify positive customer experiences.

Conclusion

In conclusion, the importance of tracking customer feedback on rating and review sites cannot be overstated for marketing directors. It is a critical tool for measuring brand health, understanding customer needs and perceptions, and driving strategic decision-making. By effectively leveraging this feedback, brands can enhance their reputation, improve customer satisfaction, and achieve sustainable growth in an increasingly competitive marketplace.

Slight easing in cost of living crisis

Fri, 23 Feb 2024 11:10:29 GMT

46% in ONS survey report increased living costs last month

The 10 most loved romcoms

Wed, 14 Feb 2024 10:25:15 GMT

Roman Holiday tops the list.

Competitive Benchmarking

Fri, 09 Feb 2024 15:59:27 GMT

Keeping ahead by keeping watch over your competitors

In today's fiercely competitive marketplace, brand managers are continuously seeking strategies to not only sustain but also enhance their brand's market position. Competitive benchmarking is valuable tool, offering insights that are crucial for keeping brands ahead of the game.

This article discusses the strategic significance of competitive benchmarking for UK brand managers, providing a comprehensive guide on its application for improved market performance.

An Introduction to Competitive Benchmarking

Competitive benchmarking involves a systematic comparison of your brand's key performance metrics against those of leading competitors within the industry. This technique is instrumental in uncovering performance gaps, identifying industry best practices and setting realistic performance targets. By understanding where your brand stands in relation to its competitors, you can make informed decisions that drive improvement and innovation.

For brand managers, the value of competitive benchmarking lies in its ability to offer a clear perspective on the brand's competitive landscape. It aids in recognising market trends, discerning areas of competitive advantage and vulnerability and pinpointing opportunities for growth and differentiation. In essence, it equips brand managers with the knowledge to make strategic decisions that propel the brand forward in the market.

A methodical approach to competitive benchmarking involves several pivotal steps:

Identifying Key Performance Indicators (KPIs): Determine the metrics that are most indicative of your brand's performance and align them with your strategic objectives.
Selecting Competitors for Benchmarking: Choose competitors that represent a broad spectrum of your market, from direct competitors to those that exemplify best practices.
Data Collection: Gather relevant data through industry reports, market surveys, text analytics and other research methods.
Analytical Evaluation: Analyse the collected data to understand your brand's performance relative to your competitors'.
Formulating Strategic Actions: Translate the insights gained into actionable strategies aimed at enhancing your brand's market position.

What metrics should you collect?

For comprehensive competitive benchmarking, brands need to collect a variety of metrics that cover different aspects of business performance, market presence and customer perception.

The actual metrics needed will be different for each business but here is a general set to get the ball rolling, along with the source for obtaining them:

Brand Awareness : Measures how familiar the target market is with a brand. Source: Surveys and polls targeting the general population or specific market segments, asking respondents if they are aware of the brand and its competitors. Brands can also use syndicated surveys such as YouGov's BrandIndex or use Google Trends as a proxy.
Market Share : Represents a brand's share of the total sales in its industry. Source: Industry reports, sales data, financial reports published by companies and market research firms.
Customer Satisfaction (CSAT): Indicates how satisfied customers are with a brand's products or services. Source: Customer surveys that ask respondents to rate their satisfaction levels post-purchase or post-interaction for the brand and competitor. Syndicated studies can also be used.
Net Promoter Score (NPS): Measures customer loyalty and the likelihood of customers to recommend a brand to others. Source: Customer surveys asking how likely they are to recommend the brand and a selection of competitors on a scale of 0 to 10. Some syndicated surveys contain measures approximating to NPS.
Likelihood to Purchase : Assesses the probability of consumers choosing a brand's product or service over its competitors. Source: Consumer surveys asking about their purchase intent regarding specific products or services.
Positive Sentiment : Assesses the positive perception and feelings towards a brand. Source: Social media analysis, customer reviews on websites, and sentiment analysis tools that evaluate the tone of online mentions and discussions.
Engagement Rate : Reflects the level of interaction and engagement consumers have with a brand's content on digital platforms. Source: Social media analytics for likes, shares, comments, and other engagement metrics on platforms like Facebook, Twitter, Instagram, and LinkedIn.
Website Traffic : Measures the volume of visitors and visits a website receives. Source: Web analytics tools like Google Analytics that track the number of visitors, page views, bounce rate, and other related metrics.
Conversion Rate : The percentage of website visitors who take a desired action, such as making a purchase or signing up for a newsletter. Source: Web analytics tools that track user actions and conversions against the total number of visitors.
Product Quality and Innovation : Assesses how a brand's products or services compare in terms of quality and innovation. Source: Product reviews and ratings on e-commerce platforms, expert reviews, patent filings (for innovation), and industry awards.

Collecting these metrics involves a mix of primary research (such as surveys and direct customer feedback) and secondary research (such as industry reports and online analytics)

Insightful Analysis

To facilitate a thorough and efficient analysis, brand managers can utilise a variety of tools and techniques. These range from SWOT analysis, which helps in assessing strengths, weaknesses, opportunities and threats, to advanced data analytics that help derive deeper insights into market trends and consumer behaviour.

Advanced data analysis techniques have significantly expanded the capabilities of competitive benchmarking, enabling brand managers to gain deeper insights into market dynamics, competitor strategies and performance gaps. Here are several advanced techniques that we can apply in the context of competitive benchmarking:

Predictive Analytics: Utilises historical data to predict future trends, customer behaviours and potential market shifts. This can help brand managers anticipate competitor moves, market demands, and identify areas where the brand can innovate or improve.
Sentiment Analysis: By analysing customer reviews, social media conversations, and other text data, sentiment analysis provides insights into public perception of different brands. This technique allows brand managers to understand consumer sentiment towards competitors, identifying strengths to emulate or weaknesses to exploit.
Social Network Analysis (SNA): This technique analyses social networks to understand the relationships and influences among individuals or entities. In competitive benchmarking, SNA can identify key influencers and opinion leaders in the market, understand competitor relationship networks, and analyse the flow of information between competitors and customers.
Cluster Analysis: This statistical technique groups objects that are similar to each other but different from objects in other groups. It can be used to segment competitors into clusters based on various attributes, such as customer demographics, product offerings, or market positioning, to identify competition gaps and opportunities.
Text Mining: Involves extracting useful information from text sources. For competitive benchmarking, text mining can analyse competitors' marketing materials, press releases and product descriptions to identify focus areas, strategic priorities and product features.

Integrating these advanced data analysis techniques into competitive benchmarking processes can provide brand managers with a more nuanced understanding of the competitive landscape. By leveraging these insights, brands can make informed strategic decisions, identify innovative opportunities, and ultimately enhance their competitive position in the market.

Conclusion

In conclusion, competitive benchmarking is a critical tool for brand managers aiming to navigate their brands towards market leadership. By systematically comparing your brand against industry leaders and innovators, you can uncover valuable insights that inform strategic decision-making and drive continuous improvement. As brand managers in the UK strive to position their brands for success in an increasingly competitive environment, embracing competitive benchmarking is not merely an option but a strategic imperative.

Polls apart

Fri, 26 Jan 2024 17:10:57 GMT

What do opinion polls tell us about the likely outcome of the election?

With a general election due within the next 12 months, we look at what opinion polls can tell us about the likely result. We ask how accurate they are as a predictor of things to come and why polls on voting intention should be treated with a large dose of salt.

If there were a general election held tomorrow, which party would you vote for?

The current opinion polls give the Labour Party a substantial lead over the conservatives and, as the graphic shows, Labour has led for two years. The graphic is based on an aggregate of all polls. The most recent YouGov poll for 23rd/24th January 2024 gives Labour a 27 percentage point lead.

Source : Electoral Calculus

Does this consistent Labour lead mean the general election is a foregone conclusion? At this point, pollsters break out in a cold sweat and remember how they got it so wrong in 1992 and again in 2015.

The predictive power of opinion polls has been analysed by Paul Whiteley, LSE . He correlated the actual results of the election with opinion polls taken 1 month, 3 months, 6 months, 12 months and 24 months away from the election, for every UK general election between 1945 and 2019 (the most recent one).

His work tells us that opinion polls have some predictive power in an election year, especially closer to the election date. However, the predictive power is not perfect, as seen in the drop in correlation over time. Whiteley found that for every time period away from the election, opinions polls were more accurate at predicting conservative share than they were Labour.

Correlations between vote intention polling and election results for Labour and the Conservatives in general elections, 1945-2019

Source : Whiteley P, How accurate are the polls when forecasting election outcomes well into the future?

From the chart, we see that:

Conservative Party: The correlation between polling and actual results tends to be relatively high and remains fairly consistent over time, with the highest correlation at 1 month out (0.9) and the lowest at 24 months out (0.57).

Labour Party: The correlation for the Labour Party is generally lower at each corresponding time point compared to the Conservative Party, with the highest correlation at 1 month out (0.84) and the lowest at 24 months out (0.4).

Why are opinion polls more accurate for Conservatives but less so for Labour?

The lower correlations for the Labour Party as compared to the Conservative Party might be due to a variety of factors:

Shifting Voter Base: The Labour Party's voter base might be more volatile or less predictable over time, possibly due to shifting political alliances, changes in the demographics of its supporters, or reactions to policy announcements.

Differential Response Rates: Labour supporters might be less likely to participate in polls or might be less forthcoming about their voting intentions, which could result in underestimations of their actual support.

Late Swings: Labour might be more susceptible to late campaign swings, either positively or negatively, making earlier polls less predictive of the final outcome.

Marginal Seats and Tactical Voting: The UK's first-past-the-post system could also play a role. If Labour contests more marginal seats where small shifts in voting patterns can lead to changes in seat allocations, this could affect the correlation between national polls and actual results.

Conclusion

The data shows that while opinion polls can provide a useful gauge of public sentiment at a given time, they are less reliable the further out from an election they are conducted.

Our advice to Sir Keir? Don't break out the bubbly just yet, there's still all to play for.

The Year Ahead for House Prices

Fri, 19 Jan 2024 14:12:12 GMT

House prices likely to flatline in 2024

Depending on what source you look at, UK house prices in 2023 were either slightly up or slightly down. Whichever way you look at it though, the market seemed to defy expectations of a huge drop in prices.

In this article we look ahead to see what might happen to house prices in 2024.

Background

During 2023 house prices either dropped year on year by 1.8% (Nationwide BS) or increased by 1.7% (Halifax HPI). These figures are based upon properties on which a mortgage was granted so differences will reflect each bank's lending criteria. The more authoritative source is the Land Registry, where sales of all houses whether by mortgage or cash are registered. The Land Registry data lags behind the Halifax and Nationwide but the most recent figures shows that UK house prices fell by 2.1% in the 12 months to November 2023. This is a far cry from the 10% many predicted as the cost of living crisis began to bight hard in 2023.

The economy

In 2024, UK house prices are at a pivotal point, influenced by a range of factors including the cost of living crisis, global economic conditions and the historically high ratio of house prices to income. The cost of living crisis, fuelled by inflation and energy costs, is straining household budgets, potentially limiting buyers' ability to afford homes. Globally, economic uncertainties, such as the aftermath of the pandemic and geopolitical tensions, may affect investment and interest rates, further impacting the housing market.

Affordability

The high price-to-income ratio, a longstanding issue in the UK housing market, continues to challenge affordability. Research by the FT calculates that average house prices are over 9 times average earnings. This compares with four times for most of the 1980s and latter part of the 1990s and is at its highest level since 1876.

While this ratio underscores the substantial growth in housing prices over the years, it also highlights the widening gap between house prices and average incomes. Data from Nationwide BS estimates that mortgages now cost 38% of disposable income (over 50% in London and the South East) and that the average deposit is equivalent to a year's salary. This disparity raises concerns about the long-term sustainability of the housing market, as it could lead to a potential market correction if prices become too detached from average earnings.

Interest rates

Interest rates have a significant impact on the UK housing market. 2023 was characterised by increased interest rates adding hundreds of pounds to mortgage payment. In 2024, if interest rates rise again, it could lead to even higher mortgage costs, potentially reducing demand for housing as borrowing becomes more expensive. This might slow down the rate of house price growth, or in some cases, lead to price declines. Ultimately, the direction and magnitude of interest rate changes will play a crucial role in shaping the UK housing market dynamics in 2024.

The general election

The impact of the general election on UK house prices in 2024 can be speculative and multifaceted. Elections often bring uncertainty, which can temporarily slow housing market activity as buyers and sellers wait for policy clarity. The election outcome could influence market confidence and economic policies, affecting interest rates, investment, and consumer spending. Manifesto promises on housing policy, taxation, and broader economic strategies could either bolster or dampen market activity. Historically, market reactions to elections vary, and the specific context of 2024's political landscape will be key in determining its impact on the housing market.

The outlook

Looking ahead, the trajectory of UK house prices in 2024 will likely be shaped by these factors. If the cost of living continues to rise and economic uncertainties persist, there could be a cooling effect on house prices. However, the market's resilience, driven by factors such as limited housing supply and the desire for homeownership, might counterbalance these pressures to some extent.

In conclusion, the UK housing market in 2024 is navigating through a complex landscape. The interplay of the cost of living crisis, global economic trends, and affordability issues will be key in determining the direction of house prices. Based on these issues, our prediction is that house prices will continue to flatline in 2024.

Plot Twist

Mon, 15 Jan 2024 10:01:31 GMT

Deriving insight from Trustpilot data

The use of ratings and review sites is a well known source of insight for brands. Previous articles, such as this one on harnessing publicly available data , have shown the valuable insight that can be derived from techniques such as sentiment analysis.

In this article we take a slightly different twist by looking at the insight that can be derived from analysing the behaviour of reviewers.

In the first part we discuss tree plots. This is a visual technique that provides insight on the buying behaviours of reviewers. In the second part, we look at how regression analysis can be used to derive drivers of loyalty.

Why is a tree plot good for insight generation?

A tree plot, also known as a tree map, is a data visualisation technique used to display hierarchical data using nested rectangles. When thinking of a tree plot, picture a big box that represents all data, divided into smaller boxes for each category, with the size of each box showing how big or important that category is. These boxes can be further split into even smaller ones to show sub-categories, letting you see at a glance how different parts of the data compare in size and importance.

When applied to customer reviews, a tree plot can provide insights into the brand's audience and their interests or buying behaviours by showing the categories that a brand's reviewers also review.

Customer Interests: By seeing the other categories that customers of a brand review, that brand can understand what other products or services their customers are interested in. This can guide cross-promotional activities, partnerships, or expansions.
Market Positioning: The brand can see how it's positioned in the market in relation to other sectors. If brand reviewers often review high-end clothing stores, for example, it might suggest that the brand appeals to a more affluent demographic.
Cross-Selling Opportunities: Understanding the other interests of their customers can help the brand identify opportunities to cross-sell related products or services.
Targeted Marketing: Insights from the tree plot can inform targeted marketing strategies. If a significant number of customers are also reviewing health and wellness services, for instance, marketing campaigns could be tailored to this interest.
Customer Segmentation: The brand can segment its customer base more effectively by using the insights to create specific profiles based on the other categories of products or services they review.
Competitive Analysis: It can provide an understanding of the competitive landscape by showing what other types of businesses attract the brand's customers.
Product Development: Insights into what other categories customers are interested in can inform new product development strategies.

Tree Plot case study method

To illustrate the insight that a tree plot can provide, we have applied the technique to reviews left on Trustpilot for supplements and sports nutrition brand Myprotein.

To generate the tree plot, we downloaded the review history of all Myprotein reviewers to identify what other brands they reviewed. From this we are able to plot the categories that they shop and their relative importance. Trustpilot uses a hierarchical system to categorise brands. Trustpilot categorises Myprotein in the Beauty & Well-being category and Personal Care and Fitness and Nutrition sub categories.

The categorisation enables comparisons to be made with direct rivals in the supplement and sports nutrition market but also more widely with brands in adjacent markets. It should be noted that the categorisations used are those recorded by Trustpilot and it is possible that some brands could be incorrectly assigned - for example JD Sports is assigned to the Events and Entertainment/Wedding & Party/Gift Shops categories. This and other examples or obviously incorrectly categorised brands were manually reclassified.

The Myprotein Tree Plot

The interactive graphic below shows the relative importance to Myprotein reviewers of each category and sub-category. The graphic is fully interactive so clicking on the tiles takes you down further to individual brands.

For Myprotein, the Shopping and Fashion category has the highest of reviews left by people who also reviewed Myprotein. Within that, the Clothing and Underwear sub category is the most important with Clothing Shops the most important with this sub category. Superdry, Next, Very and ASOS are the most commonly reviewed brands. This makes a lot of sense because as well as selling protein powders and nutrional supplements, Myprotein also sells activewear. Immediately, therefore, we see the brands that are in the m inds and shopping carts of people who buy Myprotein products.

We also see the impo rtance of less tangental markets. For example, online markets, price comparison websites and jewellery shops all feature among the reviews left by Myprotein reviewers. This is useful information because from these we can start to infer some of the characteristics of Myprotein purchasers that we are not able to get directly from Trustpilot. So for example, the use of price comparison sites suggests people that are keen on saving money. The Baby Shop tile suggests a young family demographic. In other categories we see airport car parking brands which points to people going on holidays abroad. There are categories that point to lifestyles as well as life stages. This is great for segmentation. The tree plot also identifies other brands that are important to Myprotein shoppers. This is great information for partnerships and brand associations.

It's important to remember that the data discussed above relates to Myprotein customers who leave reviews on Trustpilot. These may not necessarily be representative of Myprotein customers but should be assumed to be indicative of them.

Using Trustpilot data to identify the drivers of loyalty

We can also use Trustpilot data to identify factors that can influence loyalty. We define loyalty when a reviewer leaves more than one review for the brand with a minimum gap of ten days between reviews. We decided to leave this gap based on a widescale analysis of reviews over time, in order to cover for the situation where a person edits or adds to their initial review because there was a problem.

Using a logistical regression, we modelled the relationship between the categories that Myprotein reviewers also leave reviews for and whether they leave multiple reviews (a proxy for purchases) of Myprotein products. For the regression, we used data from all reviewers from when the brands Trustpilot account was established until late 2023. We also decided to look at the lowest hierarchy level of review category, as we believed that would gives us the most detailed analysis.

While some potentially important variables were unavailable (e.g age, income, location), we managed to generate a gender variable using a text analysis model. We accounted for factors which also have a relationship with recurring reviews of Myprotein products such as gender (men tend to more frequently leave multiple Myprotein reviews) and review frequency (people who review more frequently tend to more frequently leave multiple Myprotein reviews) by including them in the regression.

The top 5 drivers are shown in the table below. The results are expressed as odds ratios. These represent the odds that a Myprotein customer who has reviewed a product in a different category will leave multiple reviews for Myprotein, compared to the odds that a Myprotein customer will leave multiple reviews for Myprotein in the absence of a review in another category.

The table below presents the mean estimate of the odds ratio for the customers included within the regression and the range in which we would expect the true value of the odds ratio to fall 95% of the time given our regression criteria.

These odds ratios suggest that reviewers who are active in categories related to gym, fitness and nutrition, exercise equipment, activewear and health food are more likely to leave multiple reviews for Myprotein. This indicates a stronger engagement with the brand among customers interested in these lifestyle areas.

It is also important to keep in mind that there might be other intervening factors at play whether that be a variable not considered within the regression (ie age or location) or an artificial influence on the customer reviews. One such artificial influence could be a surge in reviews for a particular company as a result of a considered strategy by the company rather than a natural result of more people buying their product and leaving reviews. Thus the categories these customers also review may be subject to an unforseen time dependence, as a certain category may be particular popular at the same time as the companies strategy or the company may be trying to actively gain a customer base within a particular niche as part of its strategy. Some of this could potential be mitigated by looking at results over multiple different time periods, which is something that we would look at in a more thorough analysis.

Despite the caveats, the information derived from the analysis is valuable. Myprotein could use this information to further target and tailor their marketing efforts to these segments to encourage repeat purchases and reviews.

Overall

There are some limitations to a tree plot analysis and regression analysis derived from Trustpilot data. The major limitation is the lack of demographic data such as age and location that might explain purchase behaviour and review behaviour better. Inferences can be made based on categories purchased, but these will only ever give partial demographic insight.

Another drawback is that many reviewers only leave one review and so there may be too few reviews in some of the smaller sub categories to derive any meaningful insight. There are issues relating to mis-categorisation that need to be addressed.

And finally, given that this is just a surface level analysis rather than something more comprehensive there are still other routes for analysis such as investigating reviews over different time periods, looking into seasonality or the topics expressed in review content, to name a few.

Despite these limitations, this case study has shown the depth of insight that can be derived from an analysis of Trustpilot data and the way the information can be used to explain the market and a brands place in it, who buys from that brand and what other brands they buy. It can also give a good steer on which category buyers can be targeted and which brands are ideal for associations and partnerships.

A question of insight

Wed, 20 Dec 2023 16:11:19 GMT

Key steps in ensuring your questionnaire gives the answers you need

In the rapidly evolving world of brand engagement and marketing, understanding your audience through effective market research is paramount. A well-designed questionnaire is a fundamental tool in this process, enabling you to gather accurate and insightful data.

This guide outlines the three crucial steps and best practice in designing a market research questionnaire, emphasising the importance of thoughtful construction to achieve meaningful results.

Step 1: Define Your Objectives

Start by clearly defining what you want to achieve with your market research. Your objectives will guide every aspect of the questionnaire design, from the types of questions you ask to how you plan to analyse the results. A lot of brands use market research to cover off many issues and while it might seem good way of getting consumer insight on a broad range of topics, the danger of this scatter gun approach is that you can only ask a limited number of questions on each issue. You could miss out on detail with such a broad brush approach.

Be clear about your priorities as a business and stick to gathering information about them. remember, that you only have around 15 minutes maximum to hold respondent's interest with an online survey, so prioritising is vital.

Step 2: Choose the Right Question Types

In market research, the types of questions you choose can significantly impact the quality and usability of the data collected. Here's a more detailed look at different question types:

Closed-Ended Questions: These questions are structured to provide specific responses. They are easy to analyse and are essential for quantitative analysis. Examples include single-choice and multiple-choice questions and rating scales.

Example: "How often do you use product X?" with options 'Every day', 'most days', 'once or twice a week', 'less often than once a week','never'.
Analysis Advantage: Simplifies data collection and analysis, allowing for straightforward statistical analysis.

Open-Ended Questions: These questions allow respondents to answer in their own words, providing richer, qualitative data. They are crucial for understanding motivations, feelings, and attitudes in depth.

Example: "For what reasons do you buy product X?"
Analysis Challenge: Requires thematic analysis or coding to categorise responses, which can be time-consuming but offers deeper insights.

Rating Scales (Likert Scale): Used to assess attitudes or opinions on a continuum. These scales are effective for measuring the intensity of feelings about a particular topic.

Example: "On a scale of 1 (very dissatisfied) to 5 (very satisfied), how would you rate your satisfaction with product X?"
Analysis Nuance: Allows for nuanced understanding of attitudes and can be used in various statistical analyses, including mean scores and trend analysis.

Ranking Questions: These questions ask respondents to prioritise or rank preferences in a specific order. They are useful when you want to understand relative preferences or priorities.

Example: "Please rank the following features of product X in order of importance to you: Durability, Design, Price, Brand Reputation."
Analysis Insight: Helps in understanding the relative importance of different factors and can inform prioritization in product development or marketing strategies.

When questions go bad

Getting it wrong has serious implication for your research. Here are a few examples of bad question design and the impact it has on the overall research.

Vague or Ambiguous Questions:

Bad Example: "Do you think product X is just right for your needs?"
Why It's Bad: The term "just right" is vague and subjective, leading to varied interpretations by respondents.
Implication: Data collected will be inconsistent and unreliable, making it difficult to draw any meaningful conclusions.
Solution: Ask respondents to select words from a list that best describe their attitude to the product or use a ratings scale to measure consumer responses to attributes such as price, design, effectiveness and so on.

Double-Barrelled Questions:

Bad Example: "How satisfied are you with product X's price and quality?"
Why It's Bad: This question addresses two different aspects (price and quality) simultaneously, making it impossible for respondents to give a clear answer if their feelings about each aspect differ.
Implication: Responses will be confusing, and it will be impossible to discern which aspect (price or quality) the feedback is referring to.
Solution : Ask about each attribute in turn.

Leading Questions:

Bad Example: "Don't you agree that product X is the best in the market?"
Why It's Bad: This question is leading, as it suggests an expected response, influencing the respondent to agree.
Implication: Results will be biased towards positive feedback, failing to provide an honest assessment of the respondent's true opinions.
Solution : Use statements and ask respondents to pick which one they agree with the most. It's important to have balance, eg two positive and two negative statements with a neither/nor option in the middle.

Questions with Assumptions:

Bad Example: "Do you use product X daily, weekly or monthly?"
Why It's Bad: This question assumes the respondent uses the product daily, which may not be true.
Implication: Respondents who don't use the product daily might either skip the question or provide inaccurate responses, leading to misleading data.
Solution: Split into two questions. First question should give a list, including yours, and ask which one is used (if any), and then ask a second on frequency of use.

Complex or Technical Questions:

Bad Example: "What is your opinion on the efficacy of our product X's proprietary technology?"
Why It's Bad: The question is too technical for a general audience and assumes a level of knowledge the respondent may not have.
Implication: Respondents may be confused, leading to skipped questions or made up responses, thereby compromising the data's integrity.
Solution: Use plain language and ask respondents to assess each attribute in turn.

Step 3: Design with Analysis in Mind

This step is crucial for ensuring that the data you collect can be effectively analysed to meet your research objectives.

Align Questions with Objectives: Make sure each question directly contributes to the objectives of your research. If a question doesn’t provide data that can be used to answer your research questions, it might be unnecessary.
Consider Data Analysis Methods: Think about how you will analyse the data while writing the questions. For quantitative data, consider the statistical tests you might use (e.g. regression analysis, t-tests) and ensure your question format is compatible. For qualitative data, plan for thematic analysis or content analysis and structure questions to facilitate this.
Pre-coding Open-Ended Responses: If you plan to include open-ended questions, consider how you will categorise responses. Developing a coding scheme in advance can streamline the analysis process.
Pilot Testing for Data Analysis: Pilot testing helps not only in refining questions but also in testing your analysis plan. It can reveal if certain questions are not yielding the type of data needed or if the responses are too varied to be analysed effectively.

Conclusion

In conclusion, a well-designed questionnaire is a critical tool in market research, providing valuable insights that can shape brand and marketing strategies. By following these steps and focusing on clear, unbiased, and purpose-driven questions, brand and marketing managers can ensure that their research effectively captures the voice of their customers, leading to more informed and impactful business decisions.

Remember, the key to successful market research lies not just in collecting data, but in collecting data that is accurate, relevant, and actionable.

With a bit of time and experience, brands should be able to navigate these steps. But if time is tight and experience limited we are happy to help ensure you get what you need from research. We offer a range of research services from a free questionnaire health check to undertaking the full process for you. Please get in touch to find out more.

Essential Skills - Have You Got Them?

Fri, 01 Dec 2023 17:23:06 GMT

Evolving market research skills in the AI era

Introduction

The landscape of market research is undergoing a paradigm shift, driven by the relentless advance of artificial intelligence (AI) and data analytics.

As these technologies become integral to the field, they are redefining the skills required for effective market research. This article explores the emerging skillset market researchers will need to thrive in the future, contrasting these with the traditional skills that have dominated the field until now.

Traditional Market Research Skills

Traditionally, market research has revolved around skills like:

Survey Design and Implementation: Crafting questions that elicit valuable responses.
Focus Group Facilitation: Engaging directly with a target demographic.
Data Collection and Basic Analysis: Gathering and making sense of consumer data.
Communication and Reporting: Translating findings into actionable insights for stakeholders.

These skills have been the bedrock of market research, emphasising direct interaction with data and participants.

Emerging Skills in the AI-Driven Era

As AI and data analytics reshape the field, the skillset for market researchers is evolving:

Data Science and Analytics: Understanding and leveraging complex algorithms and data models.
AI Literacy: Comprehending how AI can be used to automate and enhance research processes.
Advanced Statistical Analysis: Moving beyond basic analysis to more sophisticated statistical techniques, often aided by AI.
Data Visualisation and Interpretation: Translating complex data sets into understandable and actionable formats.
Tech Savviness: Adapting to and utilising new tools and platforms that incorporate AI and analytics.
Ethical Considerations and Bias Recognition: Ensuring AI-driven research is conducted ethically and is free from algorithmic biases.

The Shift: From Manual to Machine-Assisted

The key difference between traditional and future skills lies in the shift from manual, direct methods to machine-assisted, indirect techniques. While traditional skills focused on human-led processes, future skills will require researchers to effectively partner with AI and analytics tools. This doesn't diminish the value of human insight but rather augments it with the speed, efficiency and depth that AI offers.

The Human Touch in an AI World

Despite the tech-centric future, the human element remains crucial. Skills in understanding human behaviour, empathy and ethical considerations are more important than ever to ensure that AI-driven market research remains grounded in human context and ethical standards.

Continuous Learning and Adaptation

The rapid evolution of AI and data analytics means that continuous learning will become a necessary skill in itself. Market researchers must be agile, willing to update their skill set, and adapt to new methodologies and technologies.

Conclusion

The future of market research is a blend of AI and human insight. As we step into this new era, market researchers must evolve, embracing new tools and techniques while retaining the core principles of understanding consumer behaviour and ethical research practices.

The fusion of traditional skills with AI and data analytics competency will define the successful market researcher of the future.

Harnessing Public Data for Business Success

Thu, 30 Nov 2023 16:00:44 GMT

Unveiling brand insights

In today's data-driven world, business decisions no longer need to wait until lengthy and costly market research has been undertaken. Instead, vast amounts of publicly available information are available at our fingertips, facilitating quick and insightful decision-making. This great source of data can provide critical insights about a brand's positioning, perception, and performance.

Here's how businesses can leverage public data sources to glean actionable brand insights.

1. Understanding the Importance of Public Data

Brands operate in an interconnected environment. Every interaction, be it with customers, competitors or stakeholders, leaves digital footprints.

These footprints, when aggregated, form a repository of public data that ranges from social media mentions to industry reports.

2. Sources of Public Data

Social Media Platforms: Platforms like Twitter and Instagram are goldmines for brand insights. Analysing mentions, hashtags, shares, and comments can provide a clear picture of the brand's sentiment, popularity and areas of concern.
Review Sites: Websites like Trustpilot, TripAdvisor and Amazon provide user-generated reviews. Analysing these can offer insights into product performance, customer pain points and areas for improvement.
Government and Industry Reports: Often overlooked, these reports provide macro insights about industry trends, regulatory changes and market positioning.
News and Media Outlets: They give a perspective on the brand's public perception and any potential PR challenges or opportunities.

3. Steps to Derive Insights

Data Aggregation: The first step is gathering data from various sources. There are several tools and platforms available that can assist in scraping or collecting relevant data points.
Data Cleaning: Raw data, especially from public sources, can be messy. Time needs to be invested in filtering out the noise to identify genuine and relevant insights.
Analysis: Utilise analytical tools to decode patterns, trends, and anomalies. Sentiment analysis can be used to gauge public sentiment or time-series analysis to understand brand performance over time.
Visualisation: Data visualisation tools can simplify complex datasets and make them comprehensible. Dashboards, graphs, and charts can help stakeholders quickly grasp the essence of the findings.

4. Transforming Insights into Action

Data, irrespective of its volume, is of no use if it doesn’t lead to actionable insights. Here are some of the things you can do to extract insight from the data:

Monitoring Brand Perception: Social media conversations, online reviews and customer feedback, offer a real-time glimpse into how consumers perceive your brand. Through sentiment analysis and topic modeling, you can identify recurring themes, positive sentiments, pain points and emerging trends. This information can guide your branding strategies, customer engagement efforts and product improvements.

Competitive Analysis: Public data can help you benchmark your brand against competitors. By analysing their online presence, customer sentiment and product performance, you can identify gaps in the market that your brand can fill. Furthermore, understanding their strengths and weaknesses can inform your differentiation strategy and help you refine your market proposition.

Tracking Market Trends: Publicly available data sources like news articles, industry reports and social media trends can be goldmines for identifying emerging market trends. By analysing these sources, you can gain early insights into shifts in consumer preferences, technological advancements and regulatory changes. Adapting your strategies in response to these trends can help you stay ahead of the curve.

Influencer and Partnership Identification: Identifying potential influencers and partnerships becomes more strategic with public data. By analysing engagement metrics and audience demographics of influencers, you can make informed decisions about collaborations that align with your brand values and target audience.

Crisis Management: Public data can serve as an early warning system for potential brand crises. By monitoring online conversations and sentiment shifts, you can detect negative narratives before they escalate. This enables proactive crisis management and damage control.

Measuring Campaign Effectiveness: Tracking metrics such as website traffic, social media engagement and online mentions, can help you measure the effectiveness of your marketing campaigns. This data can guide adjustments in real-time, optimising your strategies for better results.

5. Challenges and Precautions

While public data offers numerous benefits, it's essential to approach it with caution.

Data Privacy: Ensure that all data collection and processing activities adhere to data privacy regulations like 2018 Data Protection Act or GDPR in the EU.
Source Verification: Not all publicly available data is reliable. You need to validate the authenticity of the sources you're leveraging.
Bias Elimination: Ensure that the data analysis process is free from biases, so that insights derived are genuinely reflective of the market sentiment.

Conclusion

Leveraging publicly available data is no longer a luxury; it's a necessity for businesses aiming to thrive in the modern marketplace. The insights derived from analysing this data can shape a brand's narrative, inform strategic decisions and ultimately drive growth.

By harnessing the power of data, you unlock a world of opportunities to understand, engage and resonate with your target audience on a deeper level.

Market Understanding

Mon, 27 Nov 2023 12:04:21 GMT

Using data analytics to generate competitive insight

In an era where data is so important, understanding and leveraging it becomes pivotal for any brand aspiring to lead its market. For senior marketing executives, the strategic use of data analytics is not just a tool—it's a compass that guides decision-making, customer understanding, and innovative thinking. This quick guide explores how brands can proficiently utilise data analytics to gain a richer understanding of their markets, customers, and stay ahead in the competitive race.

Deciphering market dynamics

The market is a complex ecosystem influenced by a myriad of factors ranging from consumer behaviour to global economic shifts. Data analytics serves as a key to unlocking these complexities.

In-depth Consumer Trend Analysis: Utilising tools like Google Trends or social listening platforms, brands can monitor real-time shifts in consumer interests and discussions. This enables them to stay ahead of emerging trends. For instance, a beauty brand can track rising skincare ingredients or techniques and respond by adapting their product development pipeline accordingly.

Economic and Industry Indicator Analysis: By analysing economic indicators, industry reports, and competitor benchmarks , brands can forecast market shifts. A real estate company, for instance, might analyse housing market trends and economic indicators to predict regional demand, shaping their investment and development strategies.

Profiling and understanding customers

The cornerstone of effective marketing is a deep understanding of who the customers are. Data analytics enables a multi-dimensional view of the customer.

Advanced Customer Segmentation: Using analytical tools, brands can segment their customer base not just demographically but also based on buying behaviour, lifestyle, and digital engagement. A financial services firm, for example, could segment customers into groups like 'tech-savvy investors' or 'traditional savers', tailoring their communication and product offerings to each segment.

Behavioural Analytics and Journey Mapping: By analysing customer interactions across touchpoints, brands can map customer journeys. Analytics can provide insights into customer behaviour on websites and apps, helping brands to optimise the user experience. For a retail brand, this might mean identifying friction points in the online shopping process and addressing them to improve conversion rates.

Identifying customer needs

Predicting and responding to customer needs is a critical capability that data analytics enhances significantly.

Predictive Analytics for Demand Forecasting: Leveraging AI and machine learning, brands can predict future customer needs and preferences. A consumer electronics company, for example, might use predictive models to anticipate demand for new product features, guiding R&D investments.

Sentiment Analysis for Product Development: Sentiment analysis enable brands to gauge customer sentiment from social media and review sites. This can inform product development, as seen in a software company incorporating user feedback from online forums into their product updates.

Staying ahead of the competition

In a competitive marketplace, being informed and agile is key to staying ahead.

Competitive Intelligence Gathering: Using analytics tools, brands can track competitors’ digital strategies, from SEO to social media presence. A B2B company might analyse competitor content strategies to refine their own, ensuring they stay ahead in thought leadership.

Market Share and Performance Analysis: Data analytics can provide insights into market share, brand performance, and competitive landscape. A beverage brand could use this data to understand their position in the market, strategising to capture more share from competitors.

Overcoming challenges and ethical considerations

While data analytics offers immense potential, it comes with its set of challenges.

Data Privacy and Ethical Usage: In the wake of GDPR and other privacy regulations, brands must ensure ethical data practices. This includes transparent data collection methods and respecting consumer privacy.

Data Quality and Integration: The quality of insights depends on the quality of data. Brands must invest in suppliers that ensure data accuracy and integration across sources.

Conclusion

For marketing executives, the use of data analytics is integral to developing a nuanced understanding of the market, customers, and staying ahead of the competition. It requires a blend of technological acumen, strategic thinking, and ethical consideration. In today's data-driven world, embracing data analytics is not just an advantage—it's a necessity for market leadership.

The Seven Golden Marketing Insight Questions

Fri, 17 Nov 2023 16:51:45 GMT

Leveraging data analytics for strategic decisions

In the ever-evolving landscape of marketing, data analytics has become the linchpin for informed decision-making. Senior marketing executives, now more than ever, need to harness the power of data to unravel the complexities of consumer behaviour and market dynamics. This article delves into the seven golden insight questions that form the bedrock of strategic marketing, illustrating how data analytics can be pivotal in answering each.

1. Who are our customers?

Understanding your customer base is fundamental. Data analytics enables a deep dive into customer demographics, buying patterns, and preferences. For instance, clustering algorithms can segment customers based on purchasing behaviour, revealing insights into different customer groups. This segmentation helps in tailoring marketing messages and campaigns that resonate with each unique group.

2. What needs or problems do our customers have?

Identifying customer pain points is crucial for delivering value. Sentiment analysis, applied to social media data and customer feedback, can highlight issues or needs that your product or service can address. For example, a spike in negative sentiment about a specific problem can alert marketers to areas requiring improvement or opportunities for new product development.

3. How do our customers make purchasing decisions?

Decoding the customer's journey from awareness to purchase is key to optimising marketing strategies. Data analytics can track and analyse customer pathways, identifying touchpoints that most influence purchasing decisions. For example, a funnel analysis might reveal that certain marketing channels lead to higher conversion rates, guiding where to allocate resources for maximum impact.

4. What are our competitors doing?

Competitive analysis through data is critical for staying ahead. By analyzing competitors’ online presence, customer reviews, and pricing strategies using data scraping tools and AI algorithms, companies can benchmark their performance and identify areas for differentiation. For example, a comparative analysis of social media engagement rates can reveal competitors' strengths and weaknesses in digital marketing.

5. How is our market evolving?

Markets are not static, and neither should be your strategy. Trend analysis using time-series data can help predict market movements and evolving consumer preferences. By analysing historical data and current market trends, marketers can anticipate changes and adapt strategies accordingly. For instance, a sudden increase in online shopping in a specific demographic could indicate a shift in retail strategy.

6. What is our proposition (USP)?

Identifying and reinforcing your USP is essential in a crowded marketplace. Data analytics can help pinpoint what sets your product or service apart. Analysing customer feedback and purchase patterns can reveal what customers most value about your offering, which can be emphasised in marketing communications. For example, if data shows that fast delivery is highly valued, this can become a focal point of your USP.

7. How effective are our marketing strategies?

The effectiveness of marketing strategies is not just about intuition; it's about measurable outcomes. Advanced analytics tools can evaluate campaign performance across various channels, providing insights on engagement, conversion rates, and ROI. A/B testing can compare different marketing approaches, fine-tuning strategies for optimal results.

In conclusion, the intersection of data analytics and marketing is where strategic insights are born. By effectively answering these seven golden insight questions, senior marketing executives can steer their organisations towards data-driven success, ensuring that their strategies are not only reactive but predictive and proactive in nature. Remember, in the age of information, data is not just a resource; it’s the cornerstone of innovative marketing strategy.

Generating Insight From Text Analytics

Mon, 13 Nov 2023 11:52:08 GMT

The authentic voice of the customer

Introduction

Text analytics is increasingly becoming an important tool for brands seeking to understand the vast landscape of unstructured data. This technology enables brands to sift through large volumes of text to extract actionable insights, a task that was once insurmountable due to the sheer scale of the data involved. By utilising advanced techniques such as topic modelling and sentiment analysis, brands can gain a competitive edge through a deeper understanding of customer sentiment, market trends, and brand perception.

What is Text Analytics?

Text analytics is the process of converting unstructured text data into meaningful data for analysis, using statistical pattern learning and natural language processing (NLP). NLP techniques are applied to the text data to understand and interpret the human language within it. This includes tasks like tokenisation (breaking text into words or phrases), part-of-speech tagging, and entity recognition. The goal is to convert unstructured text into a structured format that can be analysed algorithmically.This technique allows for the extraction of valuable information from text sources, which can include customer feedback, online reviews, survey responses and website content. The insights uncovered from text analytics can inform decision-making processes, strategic planning and customer experience improvement efforts.

Applications of Text Analytics:

Sentiment analysis evaluates the emotional tone of text data, identifying whether the sentiment is positive, negative, or neutral, thereby providing a direct measure of customer sentiment towards a brand or product. This is crucial for understanding overall customer satisfaction and shaping communication strategies.

Topic Modelling applied to platforms like Trustpilot are rich sources of customer feedback. Topic modelling algorithmically uncovers recurring themes or subjects in large text datasets, enabling brands to identify dominant conversation topics among their customers. This insight is invaluable for tailoring content strategies and understanding market needs.

Aspect analysis or aspect-based sentiment analysis, takes sentiment analysis further by dissecting customer opinions about specific aspects of a product or service, such as quality, delivery or price. This granular view is essential for targeted product improvements and marketing messages.

Text classification categorises text into predefined groups, streamlining the analysis of customer feedback and identifying common issues or queries, thereby enhancing customer service efficiency and responsiveness. For example, text classification can automatically sort customer feedback into categories like 'billing', 'product issues' or 'technical support', allowing for quicker and more focused responses from customer service teams.

Trend analysis in text analytics identifies patterns and changes in customer sentiments or topics over time, offering brands foresight into evolving market trends and customer preferences. This information is key to staying ahead in a dynamic market. For example, trend analysis might reveal a growing interest in eco-friendly products among consumers, enabling a brand to adjust its product development and marketing strategies accordingly.

Customer journey analytics uses text analytics to map out the customer experience at various touchpoints, providing insights into customer satisfaction drivers and pain points. This understanding is crucial for optimising the customer journey and enhancing brand loyalty.For example, analysing customer feedback might show that users often experience difficulties during the online checkout process, indicating a critical area for improvement to enhance overall customer satisfaction.

Predictive analytics uses text data to forecast future trends and customer behaviours. This foresight enables brands to anticipate market needs, identify potential risks, and plan strategically, ensuring they remain competitive and responsive to their customer base. It involves extracting features or attributes from the text that are relevant to the predictive model. This could include the frequency of certain words or phrases, sentiment scores, topic classifications or trends over time. The model is trained using a portion of the collected data, where the outcomes are already known. For example, in predicting customer churn based on historical data where the topics customers discussed and their sentiment towards them, led to churn or retention.

Analysing open-ended survey responses : Surveys often contain open-ended questions that can yield insightful qualitative data. Text analytics can process these responses at scale, providing a quantitative lens to qualitative feedback. This approach can help brands understand the nuances of customer needs and preferences, informing product development and targeted marketing campaigns.

Website content analysis : A company's website is an important touchpoint for customers. Text analytics can evaluate the content's effectiveness by identifying the most engaging topics, the clarity of the communication and the alignment of the content with customer interests and search behaviours.

Below are some examples of the insight that can be generated through text analytics. Hover over each image for details.

Disadvantages and compromises

Despite its advantages, text analytics is not without its drawbacks. One significant challenge is the ambiguity of language—sarcasm, idioms and varying expressions can lead to misinterpretation. Additionally, the quality of insights is only as good as the data input. This means that biases in data collection can skew results. There's also the issue of privacy and ethical considerations surrounding the use of customer data. Ensuring compliance with data protection laws is paramount.

Furthermore, text analytics does not always capture the full context behind the text, potentially missing out on crucial subtleties. This is where human judgment remains indispensable, to interpret and validate the findings of text analytics.

Conclusion

For brands, text analytics offers a powerful lens through which to view the customer's feedback. It can uncover trends and sentiments that might otherwise remain hidden in the vast expanse of data. However, it should not be seen as a panacea; it is a tool that, while powerful, has limitations and requires a discerning eye to interpret its output effectively.

By understanding both its strengths and weaknesses, brands can make informed decisions that are both data-driven and intuitively guided.

Experiment AI - ChatGPT runs a focus group

Fri, 27 Oct 2023 14:06:32 GMT

Can AI do qualitative research?

Introduction

Much has been written about the potential of AI in market research but this tends to be from the perspective of quantitive research and is focused around making surveys quicker, automated and better quality.

But what about qualitative research? Does the conversational capabilities of large language models, such as ChatGPT, mean that AI could perform a transformational shift in qualitative research?

To gain some insight into the possibility, we conducted a small experiment to see whether ChatGPT could simulate a focus group environment complete with interactive conversation between participants and moderator. This article examines the process we went through and identifies some of the strengths and weaknesses of the approach.

The group discussion consisted of a short conversation between six AI generated respondents on the subject of the cost of living crisis.

Focus groups are normally around 60-90 minutes duration but as the AI generated discussion was more about proof of concept (rather than about depth and breadth of content), it lasted less than 10 minutes. Even so, there are many lessons that can be learnt from the analysis.

The video in the link above contains a clip of the discussion to help place this analysis in context. The video uses AI generated avatars to reproduce ChatGPT's output. AI video is still in its infancy and the range of avatars and voices is limited at present. Also, there are short gaps between each participant's contribution which is again down to the technology used.

Our overall conclusion is that while AI's application offers substantial efficiencies and innovative capabilities in generating insights, it simultaneously introduces challenges that cannot be overlooked.

Persona generation: achievements and limitations

The first task was to generate typical respondents. ChatGPT (version 4) was prompted to generate a range of personas that reflected a diverse personality type in relation to demographics, lifestage, attitudes and experiences. This was an iterative process using prompts to ensure that the AI 'respondents' reflected society today.

The first prompt asked for a basic persona to be generated. This generated personas such as 'The Dominator' and 'The Quiet Observer,' reflecting different communication styles and viewpoints. A second and third layer of prompts asked ChatGPT to describe a typical person who illustrates the persona and to ensure a spread of participants by age, gender and ethnicity.

This simulated diversity was designed to ensure that the conversation encompassed a variety of perspectives that might be achieved from participants drawn from the general population.

The generated personas are listed below. All the information in the panel, including names and ages, was generated by ChatGPT.

For many focus groups a more targeted audience would be required. This could be people who buy a named product, shop a particular category or use a service. No such layer was applied in this experiment due to concerns that the ChatGPT training base would have more limited data available and would therefore display a greater likelihood to generate non factual statements.

The AI simulated a reasonable variety of personas reflecting different backgrounds although with hints of stereotyping. Although it was prompted to build personas that reflect society, there is a real risk that those generated are an oversimplification or not based on fact. ChatGPT issues a warning that some of its 'facts' might not actually be true. In addition, the generated personas may not fully capture subtle cultural and emotional nuances, a gap that only real-world interactions and experiences can authentically fill.

Prompting the discussion : precision versus spontaneity

During the discussion phase prompts were used to ensure that ChatGPT elicited detailed responses from the participants and maintained a natural conversational flow. This was an interactive process with prompts added so that the AI could learn how and when the moderator should come into the discussion and how to achieve an organic conversation.

The first prompt was for the moderator to ask an open question on how each person was being affected by the cost of living crisis. This resulted in each participant speaking one after another rather than interacting as in a conversation.

ChatGPT was then prompted to rerun the conversation with more exchanges between participants and some interjections from the moderator to bring other participants into the conversation or to stop others dominating it. A third layer of prompts was used to get the moderator to pick up on interesting themes and ask the participant to expand on these.

The prompts and topic question were the only human intervention in the discussion group. All of the content was generated by ChatGPT. No prompts were used that instructed that a participant said a certain thing or raised certain issues.

One of the dangers in the use of prompts is the potential for bias. It is very easy to steer the discussion in a particular direction by prompting the model to discuss themes or for the group to take a particular stance on issues. These types of prompts were avoided in this experiment but it is recognised that prompted bias is a big concern.

Noticeable too was an observed trade-off between the AI's precision in maintaining relevance to the topic and the spontaneity inherent in human-moderated discussions. ChatGPT could not pick up on emotional undercurrents or pivot the conversation based on nuanced non-verbal cues, potentially missing out on deeper or tangential insights that a human moderator might have pursued instinctively.

Also, o ne of AI's primary drawbacks is its reliance on existing data, potentially stifling the spontaneity or unforeseen, yet valuable, tangents that human moderators might intuitively pursue. There is also a danger that the generated conversation reinforces structural bias in the training dataset, particularly relating to gender, race and a 'West' narrative.

Themes discussed, content control and authenticity

In simulating interaction, ChatGPT was directed to manage conversational flow, ensuring that each participant contributed to the discussion, thereby controlling the dominance of more assertive characters. This method was largely effective in simulating a balanced discussion environment.

The AI produced a range of themes and issues discussed among the participants. It discussed the precariousness of financial positions and gave examples relevant to the respondent's lifestage. For example, one of the respondents who rents talked about her struggle to afford sky rocketing rents. Another spoke about difficulties in feeding the family. It also generated examples of how people had changed behaviour in relation to shopping and cooking as well as spending money on non essential items. There was also an extended discussion around getting the balance right between needs and wants as well as togetherness and the sharing of resources.

The discussion also touched on the impact on mental health and wellbeing and coping strategies adopted to mitigate these impacts.

Although the discussion was broad, the balance tipped at times towards an over-controlled interaction, lacking the organic and emotional unpredictability of human conversation. The AI, could not fully replicate the empathy and psychological safety provided by a human facilitator, subtly impacting the depth and authenticity of the discussion.

The predictability of AI responses might limit the organic unpredictability and emotional depth that human discussions typically present. Moreover, the absence of genuine emotional stake in discussions could result in a lack of authentic empathy, potentially affecting the conversation's depth.

Analysing the data

After the focus group, ChatGPT was used to analyse the discussion, drawing on its ability to process and identify themes within large volumes of text. The AI provided a short report drawing together the themes and selecting quotes to illustrate points made. This part of the experiment worked reasonably well and it's easy to see how ChatGPT can be use to quickly synthesise themes and issues arising from focus groups.

Nevertheless, the model's analysis is only as good as the data it has been trained on and without experiential learning, it might misinterpret sarcasm, humour or cultural references, leading to inaccuracies in thematic interpretation. This limitation underscores the importance of human oversight in validating ChatGPT's analysis.

Conclusion

The use of ChatGPT in focus group methodologies could mark a notable advance in qualitative research, offering efficiencies in participant creation, discussion moderation and data analysis. However, these advantages come at a price. The lack of authentic human experience and emotional intuition, the potential for oversimplification and the inability to capture the full spectrum of human spontaneity and nuance present challenges that researchers must navigate. Prompt related bias is also problematic.

The AI was able to generate content that could reasonably considered in keeping with the debate around the cost of living crisis. It helps that there is a lot of information online already and so ChatGPT's database covered many points that could have been raised in a real group discussion. This in itself raises a more problematic limitation for AI focus groups - the AI draws heavily on information that already exists. For discussions around new product development or innovation, this is not always the case. The use case for AI generated discussion groups, therefore, may be limited to the synthesis of data on general topics.

It was an interesting experiment with some of it being quite impressive, particularly around the range of issues generated. That said, the technology is not in a position to replace human led qualitative research. Yet.

If you're interested in reading an analysis of the findings of the focus group then click here .

Impact of the Cost of Living Crisis on Everyday Living and Mental Health

Mon, 23 Oct 2023 16:00:50 GMT

Results from the AI focus group

Introduction

This is a report of findings of a group discussion on the cost of living crisis generated by ChatGPT. The analysis that follows is based on a short segment of the group covering the impact of the cost of living crisis on their finances , day to day activity and wellbeing.

The video above uses AI generated avatars to represent each of the participants at the discussion. An article explaining the method and the strengths and weaknesses of the experiment can be found here .

The AI focus group discussion delved deep into personal experiences amidst the current cost of living crisis. A diverse set of participants shared insights, highlighting significant themes regarding the crisis's overall impact, adaptations in everyday living, and mental health consequences.

This report synthesises the key findings, illustrated with relevant participants' quotes, to encapsulate the lived realities during this economic downturn.

The analysis that follows was written entirely by ChatGPT. The prompt used was:

Write a short report drawing out the themes from the focus group discussion of the cost of living crisis, its overall impact and the impact on everyday living and mental health. Pick out the most important themes and Illustrate key points with selective quotes and comment on how important the point is for the group. Discuss points where opinions differed. Use a professional tone, writing in paragraphs.

Precarious finances and adaptive strategies

A key theme was the heightened sense of financial precariousness compelling individuals to adopt survival strategies. Participants delineated the tangible shifts in their shopping, cooking, and general consumption habits, emphasising the necessity versus luxury balance.

Lewis captured this sentiment, stating, "It's all about what's affordable, not what's exciting." This 'back to basics' approach was a common refrain, highlighting a re-evaluation of 'needs' and 'wants.' For instance, Emily shared, "I bulk buy with flatmates now, and we rotate who cooks to save on, uh, everything..." These narratives underscore the significant recalibrations individuals undertake to navigate financial strains, pointing towards a broader theme of resilience and communal coping mechanisms.

However, opinions diverged regarding balancing economisation and maintaining a certain quality of life. Richard advocated for resilience and a proactive mindset, "It’s about navigating through, not just absorbing the impact." This perspective, while acknowledging the crisis, suggested a degree of agency, contrasting with others who felt more at the mercy of their circumstances.

Psychological impact and mental health

The psychological toll of the economic crisis emerged as a unanimous concern, cutting across all demographics. Participants reported increased anxiety, stress, and a sense of isolation, exacerbated by financial instability.

Alice’s experience was particularly telling: "The constant worry over making ends meet, providing for the kids... it keeps me up at night." Similarly, Derek noted, "It's like you're always on edge." These experiences were not isolated, indicating a pervasive atmosphere of anxiety and helplessness, with mental health directly tethered to economic pressures.

Interestingly, strategies for mental health preservation varied, with some, like Richard, emphasizing a structured routine and positive mindset, whereas others found solace in communal support or introspection. The variance suggests that mental health responses are deeply individualized, warranting diverse support mechanisms.

Communal bonds and support systems

Despite hardships, a noteworthy theme was the strengthening of communal bonds and support systems. Participants like Emily and Patricia leaned into shared resources, whether through bulk buying, sharing household responsibilities or emotional support. These narratives suggest that economic hardship, while a source of stress, also fosters communal ties and highlights the importance of social capital in resilience.

Conclusion

The focus group illuminated the multifaceted impact of the cost of living crisis on individuals from various walks of life. The economic strain has undeniably permeated all aspects of everyday living, necessitating both lifestyle adaptations and shifts in consumption patterns. More critically, the crisis's mental health impact is profound, underlining the need for comprehensive support systems.

Differing opinions, particularly regarding individual agency in handling the crisis, suggest that responses to economic hardships are influenced by personal circumstances, past experiences, and inherent coping mechanisms. These insights are invaluable, indicating that policy interventions must be multidimensional, addressing not just the financial aspect but also incorporating mental health resources and community-building initiatives.

Overall, the narratives underscore the human aspect behind the economic figures, emphasising that the ramifications of a cost of living crisis extend far beyond the wallet, influencing societal fabric and individual well-being.

Recommendations:

Implementation of community support programs focusing on mental health, providing resources accessible to individuals facing economic hardships.
Policy adjustments prioritising financial relief and sustainable economic support systems, catering to diverse needs emerging from various demographics.
Encouragement of communal coping mechanisms, potentially through government or local council initiatives that foster community resilience and mutual aid.

Ditch the tracker and other ways to save on research spend

183:752711313 (Michael Wagstaff) — Thu, 12 Oct 2023 16:15:07 GMT

With budgets tight how can brands save money and still get the insight they need?

With research budgets coming under pressure we look at six ways that brands can save money and make their research spend more cost effective.

Ditch the brand tracker

A regular brand tracking survey is a useful tool. A tracker measures awareness, helps a brand understand how it is perceived, assesses its performance on key metrics and helps evaluate the impact of marketing campaigns or new product launches.

When budgets are tight, ditching the regular tracker or at least pausing it, might be an option to consider.

When deciding whether to ditch the tracker ask yourself these questions: How much do things change from wave to wave? Are you really getting new insight every time? Is your strategy and marketing optimised as a result? Can you see a clear ROI from every wave of the tracker? Unless you can say yes to every question then ditch it, even if just for a few waves.

Of course, it has to be recognised that there are some trackers that you simply cannot ditch. An example would be one where KPI measurement is deeply embedded within the brand's business. In these cases it makes little sense to ditch it wholesale. There are two options here. First, miss a wave or two. If you track monthly then track every two months or quarterly instead. Similarly, if you track quarterly then perhaps move to six monthly. The intervals between waves will be greater but won't compromise the integrity of time series data.

Second, depending upon your budget, you might be able to do a 'light touch' wave instead of the whole thing. The key is to keep it short. You only really need to measure awareness and the things that consumers associate with your brand or product and one or two other KPIs. No more than 10 questions. That's it.

If you must do frequent or continuous brand tracking but don't have the budget for a fully bespoke one, then sign up to one of the syndicated offerings such as YouGov's BrandIndex. This will tell you most of what you need.

Use an Omnibus rather than doing bespoke survey research

Bespoke research is expensive because it has to be set up specifically for that survey. You pay all the set up costs and all the data processing and analysis costs. An Omnibus allows you to buy question space on a survey that can reach up to 2,000 people every day. With an Omnibus you share all these costs with other subscribers and usually pay on a per question basis. Some will also charge a small entry fee.

Our advice is to use an Omnibus rather than do bespoke survey research whenever you can.

Most big agencies run an Omnibus survey. Typically, they target a UK or GB population but more specialist Omnibus run in specific geographies (such as London, Manchester, Wales, Scotland) or specific groups (key workers, SMEs, financial decision makers, children and young people). Even a general Omnibus can be sliced and diced and run across multiple days so that you can get enough of the sample you need.

The point is that an Omnibus survey will meet most needs so why not see if you could save money by exploring this route?

Do it yourself

You could make savings by going down the self-serve route. Here, clients supply their own questions or use a question template provided by the supplier and run it through an online survey platform or self-service store. In relation to sample, suppliers either have their own panel of respondents or plug into the big sample providers. Results are often provided in a dashboard and some have a web based reporting tool with which you can analyse the data.

The advantage here is that you are not paying agency level fees and the survey providers are often very quick in turning things around. Some promise to get 1,000 completes by lunchtime if the survey is in field by 9am. That's pretty fast in anyone's book.

It can, however, be a drain on resources and you’re still left to sift through the data for the insight. Some providers will analyse the data for you but there is an additional fee for this.

Do online qualitative research

Online qualitative research, such as online focus groups, depth interviews, discussion forums and communities as well as mobile ethnography and diary keeping, has been around for years but really proved its value during lockdown. If you're not already doing your qual online then you should be because it's very cost effective.

Online qualitative research is cost effective because it cuts down on travel costs, expenses and incentives. It can also bring dispersed people together far more easily than in-person research can. In most cases a transcript of the issues discussed is produced in real time which not only saves money but also speeds up the process. In fact, the whole online qualitative process should be quicker than in-person research anyway.

If you're not yet ready to migrate all of your qualitative research to online then you could mix and match by doing some online and some in-person.

A word of warning though. Online qualitative research is cost effective but that doesn't mean it's cheap. It's always a good idea to shop around for quotes and consider using an independent qualitative consultant.

Make better use of existing data

Analysing existing data - your own and that from publicly available sources - can give powerful insight into how the brand is performing and what it needs to do to grow market share. It can be a better and more cost effective option than undertaking survey research.

Awareness and interaction with the brand can be measured using Google Trends. This tool is very useful because it enables a brand to see the relative volume of searches over a set time period. You can search for general terms about the brand and drill down to specific actions to see the strength of engagement consumers have.

Ratings and review sites can be mined for customer experience insight and for deriving key drivers of brand growth. Text analysis can be used to identify the attributes most associated with a brand and its competitors and how distinct it is from its rivals in the mind of the consumer.

Applying data analytics to user reviews can identify people who have moved from one brand to another and the reasons for the switch can also be identified. This is a very powerful analysis as it tells a brand how effective it is in acquiring new customers and from whom they are taking market share. It also alerts a brand to the factors that cause it to lose market share and the actions needed to retain customers.

If a brand has a presence on social media, a segmentation can be undertaken based on the accounts also followed by followers of the brand. The information provides insight on discreet groupings of followers and on the people, events and media content that interests them. This information will help with more effective targeting of existing and potential users of the brand.

A brand will have a lot of data internally that can be mined for insight. Sales, CRM and other internal data and even historic brand tracking data, can be used to undertake predictive modelling, market mapping, marketing ROI, forecasting and to construct churn and retention ratios.

Small, independent agencies is where the value's at

There's a temptation to stick with what you know and place your research with well-known market research agency. But agencies have big overheads to cover and do not always provide the best value as a result.

Outside of the traditional agencies, there is some top-notch talent working on their own or in small consultancies. These are usually people, with director level and above experience, who are dedicated to helping their clients succeed.

So, if you're looking for first-class insight with unmatched value then why not consider placing your research with an independent consultant, micro business or small agency? You'll find experts who really understand your business and are keen to get their hands dirty by doing the work rather than passing over to a junior colleague. And they don't charge agency level fees either. This means that you'll be able to get more insight for your money.

With insight budgets getting smaller, going down the smaller agency route makes perfect sense.

Takeaways

With the cost of living crisis and resultant fall in consumer spending, research budgets are being squeezed. But that doesn't mean that insight has to be put on the back burner. We have identified six ways that research and insight can be undertaken on a reduced budget.

It's inevitable that each involves some sort of trade off between what a brand ideally wants and what it can actually afford. But we would argue strongly that it's better to get most of what you need at a price you can afford rather than compromising the business through lack of insight.

Is Stoptober running out of puff?

Wed, 04 Oct 2023 16:01:15 GMT

With smoking among 25-34s rising, does Stoptober actually make a difference?

The government has signalled its intention to raise the minimum age at which smoking is legal by one year, every year. The intention is that no child turning 14 or younger this year will ever be legally allowed to smoke.

It is estimated that about 13% of the adult population of the UK smoke, which equates to around 6.4 million people. Most age groups have experienced a decline in smoking (particularly the 18-25 year group) but the prevalence among those those aged 25-34 went up in 2022. Adults aged 25 to 34 years have the highest proportion of current smokers (16%, around 1.4 million people), compared with any other age group. This is an increase in comparison with the same group in 2021 (16%, around 1.3 million people).

One of the many initiatives targeted at getting smokers to quit is Public Health England's annual Stoptober intiative. First introduced in October 2012, Stoptober is a challenge for smokers to quit for 28 days. Research suggests that if a smoker can make it that far without a cigarette they are five times more likely to quit for good. The campaign has an app that people can download for support and advice, a large social media presence to encourage a sense of mass participation and personal quit plans to help navigate through the 28 days.

So does Stoptober actually work?

The impact of the campaign

What then has been the impact of the campaign?

Public Health England (PHE) has evaluated the campaign in the past but has not published a review since 2020. This most recently evaluation shows that, 12% of smokers made an attempt to quit and one third of these (4% of all smokers) made it through to the 28 day mark.

It is not possible to compare the impact in previous years because PHE changed survey suppliers in 2019 which led to a marked difference in the scores on each metric. It is possible to say, however, that the proportion of smokers making an attempt to quit was more or less the same between 2015 and 2019 (although at a much higher level than 2020's figure). So too was the proportion who made it to 28 days (again at a higher level than 2020).

There is some evidence to suggest that awareness of Stoptober has been diminishing. Looking at Google Trends data we can see that searches for Stoptober have fallen over time. The peak search time is usually the last week in September. The figure below shows that the relative number of searches for Stoptober has declined since September 2018 with searches for 2023 in the peak week down to around one fifth of the 2018 number.

Popularity of searches for Stoptober on Google 2018-2022

Source : Google Trends. Numbers represent search interest relative to the highest point on the chart within the time period (23rd Sept 2018 to 3rd Oct 2023). A value of 100 is the peak popularity for the term. A value of 50 means that the term was half as popular. Searches peak in the last week of September each year.

The 2020 campaign evaluation reported that awareness of the campaign was down on the previous year (40% down from 54%). PHE stated this could partly be due to confusion with the Better Health campaign which also encourages people to quit smoking. There is some evidence to support this : in 2020 the volume of searches for 'Better Health' peaked in August. This seems to be a one off because in 2021 and 2022 searches peaked in January.

The question is does the plateauing of quit attempts, decline in campaign recall and drop in Google searches mean that Stoptober has run out of puff? Does it actually lead to long term behaviour change?

Does Stoptober lead to a long term behaviour change?

Stoptober encourages people to attempt to quit smoking. It is estimated that it has helped over 2.3m smokers make an attempt to quit since its introduction in 2012. That's an impressive figure. But giving up smoking is not a yes/no decision. There are many points in the journey from someone thinking about quitting to actually doing it.

Initiatives like Stoptober have an important role to play in moving people along the path of behaviour change. There are three points in particular where Stoptober has an impact: (i) motivating people who are open to attempting to quit to find out more about the help and support available (ii) providing support, encouragement and advice to those planning to quit so that they make the step to actually attempting it and (iii) encouraging those who have tried to give up in the past (New Year's resolution for example) to try again.

The focus on Stoptober being a mass participation event (and its use of social media to underpin this) is important for people who are contemplating quitting. The feeling of not being alone, combined with its emphasis on the positive benefits of quitting (rather than just on the harm caused by smoking) brings positive encouragement to people open to the idea of quitting. Things like the app, personalised quit plans and making quitting more achievable by doing it for 28 days are important in encouraging people to take action because they provide the tools necessary to move from thinking about it to actually doing something.

The biggest challenge for Stoptober is to help as many people give up for 28 days and beyond. PHE's own research shows that two-thirds don't make it to 28 days and in fact half drop out after two weeks. There will be a number of reasons for this: lack of engagement with the support tools, low confidence to succeed, diminished willpower, motivational messages not cutting through or factors such as stress and personal circumstances. For many the cravings will just be too much - smoking is an addiction after all. How can the two-thirds be better supported in the future?

The government's ambition is for England to be designated smokefree by 2030. This is defined as meaning only 5% of the population smoke. A campaign, with an appropriate level of funding, that encourages and supports people to quit can be a useful part of the making smoking obsolete toolkit.

Whether that campaign is one with a month long focus such as Stoptober or one that is part of a more holistic and ongoing campaign around better health outcomes, is up for debate.

Getting the right blend

Fri, 29 Sep 2023 15:24:22 GMT

Merging survey data with business intelligence

In the modern era of data-driven decision making, businesses have access to a wide variety of data sources. While each type of data offers its own unique insights, the magic often lies in the blending of various data types.

One such powerful combination is the amalgamation of survey data with a brand's existing metrics like sales, revenue, CRM and customer experience data. This blending can provide an all round view of brand performance and help drive strategic actions.

Why Blend Survey Data with Existing Brand Metrics?

Survey data can tell you how your customers feel, while sales and revenue data can indicate how well the brand is doing financially.

Combining these can help pinpoint areas where customer sentiments and sales diverge, providing key areas of focus. In summary, the additional insights achieved by blending survey data with your own business metrics include:

Granular Customer Segmentation: CRM and customer data can help you understand who your customers are. When this is layered with survey responses, you can gain insights into the preferences, pain points and sentiments of specific customer segments.

Predictive Analytics: Combining behavioural data with attitudinal data (from surveys) can help brands predict future trends, customer behaviours and even potential risks or opportunities in the market.

Mapping Customer Satisfaction to Sales Data: Imagine you've just launched a new product. While sales data might show robust numbers, your customer satisfaction surveys could reveal that many customers aren't happy with a specific feature. By blending the data, you identify a potential risk early on, allowing for timely product tweaks.

Lifetime Value and Feedback Correlation: By integrating CRM data with survey feedback, brands can identify if their highest-value customers are the most satisfied. If not, targeted campaigns or personalised offers can be designed to enhance their experience.

Tailored Marketing Campaigns: Let’s assume a segment of your customers, identified from CRM data, consistently purchases a particular product. If survey data shows this group desires more sustainable options, the brand can launch a marketing campaign highlighting the eco-friendly attributes of the product.

Churn Prediction: If revenue data shows a decline in repeat purchases and survey data highlights growing dissatisfaction with customer service, these combined insights can lead to proactive interventions to reduce churn.

Blending the data

It's much easier to blend data sets that I have unifying variable such as email address, customer number or postcode. But unfortunately that is not always the case.

When two disparate data sets lack a common linking variable, integrating these sources is inherently challenging but not impossible. Strategic approaches are essential to ensure that blending such data yields meaningful, actionable insights.

Below are several methodologies and strategies to consider.

1. Aggregation & Segmentation:

Aggregate both the survey and internal business data based on common dimensions such as demographics or geographical location, creating segments that are comparable across both datasets. This can involve analysing trends, patterns and average values within specific segments rather than individual level data.

2. Propensity Score Matching:

Propensity Score Matching (PSM) is a statistical technique that can be employed to make inferences in the absence of a unifying variable. It involves creating comparable groups based on observed characteristics, allowing insights to be drawn by comparing the propensity scores of the survey data to the business data.

3. Canonical Correlation Analysis:

Canonical Correlation Analysis (CCA) can be used to understand the relationships between multivariate datasets when there is no common variable. This statistical method helps in identifying patterns and correlations between sets of variables in different datasets, thereby facilitating the combination of insights derived from both data sources.

4. Data Fusion & Feature Engineering:

Data fusion techniques involve creating new variables or features that are common to both datasets, facilitating more robust analysis. Feature engineering can include creating bins or categories that can be compared or finding proxy variables that can serve as a link between datasets.

5. Advanced Machine Learning Techniques:

Leverage advanced machine learning models that can handle heterogeneous data types and are capable of finding complex, non-linear relationships between datasets, even in the absence of a clear linking variable. Techniques such as clustering and principal component analysis can help in uncovering latent structures and relationships in the data.

6. Expert Judgment & Domain Knowledge:

Consult with domain experts who have in-depth knowledge of the subject matter and industry to formulate assumptions and hypotheses about potential relationships between the datasets. Expert input can help in contextualizing the findings and prioritizing which correlations or patterns are most meaningful and actionable.

7. Use of Proxy Variables:

Identifying proxy variables that are available in both datasets is crucial. While they may not directly correspond, they can offer an indirect link. For instance, socioeconomic status could be a proxy variable linking survey responses to purchasing patterns.

8. Trend Analysis:

Compare the trends observed in both datasets over time. While it won’t offer a direct link, it could provide valuable insights into how changes in one dataset correspond with shifts in the other, allowing for deeper understanding of the dynamics between customer feedback and purchasing behavior.

9. Qualitative Analysis:

Implementing qualitative analysis methods such as thematic analysis on open-ended survey responses can also provide insights that can be related back to existing business data, enabling the identification of overarching themes or patterns that can guide business strategy.

Our expert team is able to apply these methods to your datasets. Contact us for details.

Conclusion

Blending disparate data sources without a unifying variable is a complex endeavour but one that can yield significant insights when approached strategically. Employing advanced statistical, machine learning methods, and qualitative analysis coupled with expert judgment can unearth correlations, patterns, and trends between the datasets.

The integration of these diversified approaches ensures that even without a direct link, a harmonious synthesis of data is achievable, empowering brands with enriched, multi-faceted insights to drive informed decision-making and strategy optimisation.

Reporting that maximises insight

Fri, 22 Sep 2023 10:13:51 GMT

Best Practices in analysing and reporting survey data

Market research surveys are a valuable tool for businesses to understand their customers, test hypotheses and make informed decisions. However, the full value of this tool can only be realised when the survey data is analysed and reported correctly.

In this article we list six simple steps to ensure that survey data is both accurate and impactful.

1. Checking Data for Accuracy

Data Cleaning: Before diving into analysis, ensure your dataset is clean. Remove or correct any outliers, duplicates or inconsistencies.
Quality Control: Look for patterns indicating biased or skewed responses, such as participants selecting the same answer choice for all questions. Such responses may need to be excluded.
Verify Sampling Methodology: Ensure the sample used is representative of the population you intend to infer about.
Weighting the data : Apply weights if necessary so that the data are representative. Be wary of applying large weights to small samples as this will magnify any undue influence.
If you're using an agency to do your survey make sure you discuss these steps in advance of the survey going out.

2. Computing New Variables

Creating Indices: Combine multiple variables into an index to capture broader constructs (e.g., customer satisfaction based on various service metrics).
Segmentation Variables: Compute variables that segment the audience into meaningful groups. For instance, use age and income brackets.

3. Making Inferences

Statistical Significance: Always check if observed differences or trends in the data are statistically significant, especially when comparing groups.
Causation vs. Correlation: Be cautious not to assume causation from correlations. Just because two variables move together doesn’t mean one caused the other.
Multivariate Analyses: Sometimes, considering multiple variables simultaneously (e.g., through regression analysis) can provide more nuanced insights than univariate analyses.

4. Drawing Out Insight

Ask the 'So What?' Question: After identifying a trend or pattern, delve deeper to understand its implication for the business.
Synthesise with other data sources: Blend survey data with other datasets (like sales or website analytics) for richer insights.
Benchmarking: Compare your results with industry benchmarks or past data from your own surveys to provide context.

5. Presenting Data for Maximum Impact

Use Visual Aids: Graphs and charts often convey trends and patterns more effectively than tables or text.
Executive Summaries: Start your report with a concise summary of the most crucial findings and their implications.
Tell a Story: Structure your report as a narrative, guiding the reader through the most relevant points in a logical sequence.
Avoid Jargon: Make sure your report is understandable to all stakeholders, not just those well-versed in research or your industry.
Interactive Dashboards: Tools like Tableau or Power BI can help you create interactive dashboards. Stakeholders can dive into the data at their desired depth.

6. Feedback Loop

Solicit Feedback: After presenting your findings, gather feedback from stakeholders to understand which insights were most valuable and why.
Iterate: Use feedback to refine your next survey or the way you analyse and present data.

In conclusion, market research survey data has the potential to offer significant value to businesses, but only when approached with a rigorous and systematic analysis and reporting methodology.

By adhering to these best practices, you ensure that your insights are accurate, meaningful and actionable.

Asking the Questions

Fri, 15 Sep 2023 11:17:21 GMT

Concrete Majority

Thu, 14 Sep 2023 14:33:34 GMT

How vulnerable are MPs in RAAC seats?

At the end of August, the government released a list of 147 schools in England where crumbling concrete posed a risk to the buildings. These schools were built using reinforced autoclave aerated concrete (RAAC) and those listed were in the most immediate need of repair.

What impact might the location of these schools have on an election, probably next year?

Of the 147 affected schools, three-quarters are currently held by the Conservatives. Many of these seats have large majorities but three are marginals, where the winner's share of the vote is less than 10% of that of the second placed candidate. These includes Ehser and Walton, currently held by Dominic Raab.

A further 10 seats have a majority of less than 10,000 votes and could also be considered vulnerable.

Of course, just because there is a RAAC school in the area doesn't necessarily mean that the constituents will vote against the sitting MP - and it's worth remembering that the problem predates the current government. It does, however, act to focus the minds of voters on the state of public services in their local area.

And with Labour continuing to be well ahead in the polls, some of those big majority Tory seats could come into play.

Hover over the interactive map to highlight details of the constituency and the MPs majority.

How reliable is an online survey?

Thu, 07 Sep 2023 16:30:04 GMT

Getting the balance between speed and accuracy is vital

In the age of digitalisation, the allure of online surveys for businesse, especially the DIY platforms, is undeniable. These provide a way to quickly gather data, reach a global audience and do so in a manner that is often more cost-effective than traditional methods like face-to-face or postal surveys.

Beyond ease of use, the real-time data collection allows businesses to respond swiftly to insights. Moreover, the multimedia capabilities of online platforms make them versatile, letting researchers craft surveys that are both engaging and varied.

However, with these advantages come certain concerns.

One of the most pronounced challenges with online surveys is the potential for self-selection bias. Respondents who voluntarily participate might already have a vested interest in the topic, which could skew the results. Furthermore, the so-called 'digital divide' remains a real issue; not everyone has access to the internet, which might inadvertently exclude specific demographic segments from participating. There's also the question of sincerity and authenticity. The veil of online anonymity sometimes encourages non-genuine responses. And of course, in a world flooded with online content, survey fatigue is a genuine concern; respondents can easily feel overwhelmed or indifferent.

Can the accuracy of an online survey be measured?

Statistical measures of survey accuracy are often applied to online surveys. Strictly speaking applying measures such as sampling error to an online survey is mistaken as most online surveys use what is known as a non-probability sample. In a probability survey, every person in the population has a known, non-zero chance of being selected. This allows for direct generalisations from the sample to the population. Standard errors, confidence intervals and significance tests are straightforward to compute.

For non-probability surveys, not all people in the population have a known or consistent chance of being selected. Online surveys often sample through a panel of members who have signed up to do surveys and these might not be representative of the larger population. Because of this, traditional measures of accuracy like margins of error don't have the same straightforward interpretability.

So what can be done to improve the reliability of online surveys?

There are a number of things that can be undertaken pre and post survey that will improve the reliability of the results.

Pre survey activities include:

Preventing multiple entries from the same individual is vital. Tools like cookies or IP tracking, used within the confines of privacy regulations, can help in maintaining the survey's integrity.
Before full launch, it's advisable to pre-test the survey on a small group. This can help in ironing out ambiguities and ensure the survey is free from technical hitches.
To ensure that respondents are genuinely engaged and not just skimming through, attention checks can be embedded in surveys. Such questions require careful reading and can filter out bots or inattentive participants. Along the same lines, ensuring the anonymity and privacy of respondents can boost honest participation. Being transparent about data usage and storage protocols, can foster trust.
Survey design optimisation is paramount. A concise survey reduces respondent fatigue and clear, simple language aids comprehension. Adaptive questioning, where the survey evolves based on prior responses, can also keep participants intrigued. We will cover question in the next article in this series.

Once the results are in, we suggest doing the some or all of the following:

Weighting: One common practice is to weight survey results so they align more closely with the known population on key demographic or other variables. However, even with effective weighting, there might still be unknown biases in the results because certain groups might be underrepresented.
Bootstrap Methods: Resampling techniques like bootstrapping can be used to estimate variability in non-probability samples. By repeatedly drawing samples (with replacement) from the observed data and calculating the desired statistic for each sample, you can get an empirical distribution of the statistic and its variability.
Model-Based Approaches: Instead of drawing inference from the design of the survey, a model-based approach assumes that a correctly specified model can adjust for the biases in non-probability samples. For example, post-stratification is a technique where responses are adjusted based on known population totals in subgroups.
Benchmarking: If there are other, more reliable data sources available (e.g., government statistics or probability surveys on similar topics), non-probability survey results can be compared or benchmarked against these. If they align well, it increases confidence in the non-probability results.

One final thing to keep in mind is that when reporting results from online surveys, it's essential to be transparent about the method, including how participants were recruited, potential sources of bias and any steps taken to adjust or weight the data.

In conclusion, while online survey samples present unique challenges, their potential as a data collection tool is vast. By understanding the limitations and proactively enhancing accuracy, they can serve as invaluable assets for businesses and researchers alike

The Art of Text Analytics

Fri, 25 Aug 2023 09:54:33 GMT

User reviews contain so much more than just a star rating. We applied text analytics to TripAdvisor reviews of four popular London galleries to get a more complete picture.

How EasyJet Took Off

Fri, 18 Aug 2023 15:29:53 GMT

In the latest of our How Did They Do That series we look at EasyJet

AI in Survey Based Market Research

Mon, 07 Aug 2023 07:10:47 GMT

AI is here already and there's no need to be afraid

Articles on artificial Intelligence (AI) are commonplace at the moment. And one of the really interesting things is the number that are illustrated with a robot in human form or a half robot - half human hybrid. In fact, rather like the illustration we have used for this article!

AI is often associated with a post apocalyptic world of sentient robots ruling over a dystopian landscape. This just goes to show how people associate AI with sentient beings that can act and think for themselves and unless we ACT NOW, will take over the world.

The reality is a lot more mundane. In fact AI is already deeply embedded in the survey process, particularly in relation to sampling and analysis, in a very non threatening way. Here's how.

Sampling Optimisation : AI can help in designing effective sampling strategies that ensure a more representative and unbiased sample. Machine learning algorithms can analyse historical survey data to determine the best sampling techniques that can yield the most informative results.

Improved Representativeness : With AI, it is possible to improve participant response by analysing demographic data, past survey responses, and other relevant data. This ensures that the surveys reach the most relevant audience, thus improving the quality of the sample.

Bias Reduction: AI can help in identifying and reducing various kinds of biases that can creep into surveys, such as selection bias, nonresponse bias and coverage bias. By identifying patterns in who is not responding or how certain groups are over- or under-represented, adjustments can be made to reduce these biases.

Data Cleaning: Machine learning algorithms can help in cleaning the survey data by identifying and handling outliers, missing values and other anomalies that can skew the results.

Adaptive Survey Design: AI can help design adaptive surveys where the questions posed to the respondents are dynamic and change based on their previous answers. This not only increases the relevance of the survey but also improves response rates and the overall quality of the sample.

Response Quality Check: AI can assess the quality of the responses in real-time, checking for inconsistencies or indications that the respondent is not providing thoughtful answers (for example, always choosing the same option). It can then flag these for review or potentially discard them to maintain the integrity of the sample.

Automated Follow-ups: AI can automate the process of following up with respondents who have not completed the survey, increasing the chances of obtaining a complete and representative sample.

Predictive Analysis: AI can analyse the survey data and predict potential trends or patterns in the responses, allowing researchers to modify the survey in real-time to better meet their objectives. This is particularly helpful during a soft launch phase.

Enhanced Analysis: Post-survey, AI can assist in data analysis, employing advanced techniques such as sentiment analysis, topic modelling and clustering to deliver more nuanced insights from the survey data.

None of the above points are things to fear. In fact, all of them are being done right now in most market research agencies. AI is being used to make the survey process smoother, better and more accurate.

AI is taking over the world but in a good way.

Top 5 Movies Based on a Toy or Game

Fri, 28 Jul 2023 15:47:11 GMT

IMDB ratings for films based on an existing toy or game

Fairground Attraction

Sat, 22 Jul 2023 09:28:35 GMT

What do visitors think about AltonTowers, Blackpool Pleasure Beach and Thorpe Park?

Minecraft : How did they do that?

Fri, 14 Jul 2023 13:46:55 GMT

The Rise of Minecraft: Business strategy and transferable Lessons

Introduction

In 2009, Minecraft, a video game created by a lone Swedish developer, Markus Persson, was released. It was released initially as development version that was still being tested.

Originating as a passion project, Minecraft was the brainchild of Persson's love for Lego and his desire to create a game that offered endless possibilities. The game allows players to construct and explore virtual landscapes using textured cubes or "blocks". Despite its simplistic graphics and lack of a traditional narrative, Minecraft's open-world, user-generated content model struck a chord with gamers worldwide.

This small independent game has evolved into a huge global hit, becoming one of the best-selling video games of all time. The journey from a small indie title to a multi-billion pound brand was not without its hurdles. Initially, Minecraft faced a dual challenge: a lack of awareness in an oversaturated market and a unique game design that contradicted mainstream gaming conventions.

However, Minecraft used these perceived weaknesses as strengths. It leaned into its unique identity, enticing players with the promise of unfettered creativity. The inherent freedom and simplicity of Minecraft were innovative and offered a refreshing alternative to the norm, which turned out to be its main selling points.

Navigating rough terrain

Minecraft faced considerable scepticism initially due to its departure from traditional game aesthetics and structure. Many industry experts doubted whether a game with rudimentary graphics and no set objectives could sustain interest in a market dominated by high-definition graphics and tightly scripted narratives.

Minecraft's seemingly contradictory approach to game design was not its only challenge. The game was developed by Mojang, a small and relatively unknown studio with limited resources. It was distributed for free which meant that there was a lack of funding for traditional marketing activities. Instead, Minecraft relied heavily on word-of-mouth and the growing influence of online gaming communities.

The power of community engagement

In an alignment with emerging trends, Minecraft's release coincided with the rise of platforms like YouTube and Twitch, where gamers shared their experiences and creations. Players became the game's biggest promoters, with many creating tutorials, virtual tours, and "Let's Play" videos that showcased Minecraft's unique gameplay. This organic, user-driven promotion effectively served as free marketing and was instrumental in Minecraft's viral spread.

Recognising this, Mojang adopted a marketing strategy that leveraged these platforms and fostered a close relationship with its player base. They embraced the game's modding community, encouraging players to modify and expand the game's features, effectively turning the game into an ever-evolving platform.

This commitment to community engagement and user-generated content became a cornerstone of Minecraft's marketing approach and a significant factor in its enduring popularity.

Marketing and evolution over time

Minecraft's marketing approach was as unique as the game itself. The initial marketing was largely word-of-mouth, fuelled by an enthusiastic online community of early adopters. This grass-roots approach was supplemented by a public beta phase, which allowed players to try the game while it was still under development.

As the game's popularity grew, so did its marketing efforts. Mojang began to host Minecraft conventions, known as MineCon, which further solidified the game's community.

In 2014 Microsoft acquired Mojang and the Minecraft IP for $2.5 billion. After the acquisition, the marketing strategy underwent a significant transformation. Microsoft leveraged its expansive ecosystem to promote Minecraft, bundling the game with Xbox consoles and featuring it prominently in the Windows Store.

In recent years, Minecraft has expanded into different media and merchandise, including toys, clothing, and books. Collaborations with high-profile brands have also been a successful marketing strategy. One notable example is the Burberry x Minecraft collaboration, which involved an in-game adventure, a capsule collection, and real-world experiences, each filled with creativity, exploration, and self-expression. The Burberry x Minecraft capsule collection fused in-game motifs with Burberry icons, showcasing how Minecraft's brand can seamlessly blend with other industries.

Another notable brand partnership is with Lego, the brand that inspired the game in the first place. Lego now make Minecraft related sets. The partnership has leveraged their shared ethos of creativity and open-ended play. This strategic alliance allowed Minecraft to extend its market presence through a tangible product range, expand its narrative universe and tap into new revenue streams from Minecraft-themed Lego set sales. By aligning with a globally trusted brand like Lego, Minecraft gained an increased reputation, especially among parents, potentially attracting new players. This blend of physical and virtual realms of play has not only benefited both companies financially but also enhanced the value proposition for their worldwide user communities.

Transferable lessons from Minecraft's success and how research and analytics can help

Minecraft's journey offers valuable lessons for other brands aiming to carve their path in a crowded market.

Embrace your uniqueness: Minecraft succeeded by defying the gaming conventions of its time. Brands should not shy away from their unique traits, as these can become their defining features and selling points. Market research can uncover unique selling points and differentiators that a brand possesses by analysing customer sentiments and opinions. Data analytics can help identify unique patterns in behaviour and preferences that can be highlighted and amplified to distinguish the brand in the market
Cultivate a community: Minecraft's success was fuelled by an engaged community of players. Brands should focus on building and nurturing a community that can advocate for their product or service. Analytics can identify the most engaged and influential customers within a brand's demographic, aiding in the targeted building of a community. Market research can help understand the values and interests of the community, helping the brand to engage with them on a deeper level
Leverage user-generated content: Minecraft leveraged the power of user-generated content long before it became a common marketing strategy. Brands should consider how they can empower their users to become ambassadors for their products. By analysing user feedback, brands can gain insights into their customers' perspectives and experiences, enabling them to leverage this content in their marketing strategy.
Constant innovation and adaptability: Minecraft stayed relevant by continually innovating and adapting to changes in the market, whether by introducing new game modes, collaborating with other franchises or even entering the education sector. Brands need to remain flexible and adapt their strategy based on changing market conditions and customer needs. Through market research, brands can stay on top of trends and consumer needs, enabling them to innovate effectively. Data analytics can provide real-time feedback on product performance and customer engagement, enabling rapid adaptability to market changes.

In conclusion, Minecraft's success story is a great example of how a unique idea, coupled with effective marketing and a strong community, can propel a product to unprecedented heights. It stands as a lesson that creativity, community engagement, and adaptability are important traits in the ever-evolving business world.

The graduate gender pay gap

Fri, 30 Jun 2023 14:00:35 GMT

Why are female graduates paid less than males?

Many young people in England will now have finished their A levels and will be hoping to go to university.

Going to university has long been viewed as a route to a better standard of living. The latest data from the government's Graduate Labour Market Statistics (GLMS) shows that people of working age in England (ages 16-64) who have a degree earn around £11,500 per year more than non graduates. That advantage diminishes once student loan repayments are taken into account but even so, graduates will be paid more than non graduates over their working lifetime.

The gap between graduates and non graduates has been fairly consistent over the last few years. Also consistent, however, is the salary gap between male and female graduates.

The graduate gender gap

GLMS data for the calendar year 2022 shows that there is a gap of £9,000 in the median salary of working age male graduates compared with working age female graduates. Some of this difference can be explained by part time working and the imbalance in child care duties between women and men.

But this doesn't wholly explain why a gender pay difference exists for younger graduates, aged 21-30, for whom child care and part time working may have less of an impact (the average age of a first time mother is just over 29).

The graphic below shows the difference in salaries for male graduates aged 21-30 and female graduates of the same age. It highlights a long established pay gap between younger male and female graduates.

Median annual gross salary for male and female graduates aged 21 to 30, 2012 to 2022

Source: Graduate Labour Market Statistics, June 2023. Figures are not adjusted for inflation.

Female graduates aged 21-30 have consistently earned less than male graduates of the same age. Between 2015 to 2019, the gap increased steadily year on year to £4,000. It came down sharply in 2020 but increased again to £3,000 in 2021 before coming down slightly to £2,500 in 2022.

Why do younger female graduates currently earn less than their male counterparts?

As noted above, part time hours and an imbalance in child care responsibilities are important factors but they not the only ones. R esearch in 202 1 by the Institute for Fiscal Studies identified that differences in degree choices also help to explain the gender pay gap, particularly early in the career of a young graduate. The IFS research found that women are overrepresented in degree subjects, such as psychology, creative arts, English, education and social care, that have lower financial returns.

What is not clear though, is what is driving lower salaries in jobs that psychology, creative arts, English, education and social care graduates go into. Is it something to do with the value society places on these jobs and therefore they pay less? Or is it because these jobs are mainly taken up by women and therefore they pay less? In other words, is discrimination a factor?

Looking at median salaries by industry type for graduates aged up to 30, it's clear that the gender gap runs across a wide range of industrial sectors.

Median salaries by industry for male and female graduates aged 21-30, England, 2022

Source : GLMS, June 2023. Salaries are rounded to nearest £500.

Younger male graduates in Manufacturing are paid, on average, £6,000 per year more than females, £5,500 more in Transport and Communications and £4,500 more in Banking. Female graduates earn more in Energy & Water, Distribution,Hotels & Restaurants, and Other Services but the gaps tend to be lower. In addition, the median salaries for females in Distribution, Hotels & Restaurants is the lowest for females in any sector.

The IFS research identified having children along with adversity to risk, lower propensity to bargain over salary or to ask for promotion and a lack of recognition for a woman's contribution to group work as factors that lead to female graduates earning less than male graduates. Discrimination could also be a factor.

Over the last few years gender pay gap reporting has helped to reduce the gap between men and women in the workplace. But the GLMS data shows that younger female graduates can still expect to be paid less than their male counterparts.

This is simply wrong.

Take a Kodak Moment Video

Thu, 15 Jun 2023 14:31:14 GMT

How did Kodak go from dominating the photography industry to bankruptcy?

Bouncebackability

Fri, 02 Jun 2023 14:39:28 GMT

New data from Sport England shows how some sports have bounced back from Covid dip.

Labelled with Love : Does calorie labelling help?

183:752711313 (Michael Wagstaff) — Wed, 10 May 2023 09:38:43 GMT

Does displaying calories on menu items make a difference? We look at the evidence.

It's been a year since large restaurants and takeaways were legally required to display the calorie content of meals on the menu. This rule was introduced as part of the government’s strategy for dealing with obesity. The thinking behind it was that if you know how many calories are in the meal you might consume less and eat a healthier diet.

Now one year later the big question is: has it work?

As yet the government has not published any research on the impact of the first year of the legislation. Instead, we've looked at analysis from around the world of the experience after the introduction of calorie labelling.

Research has shown that calorie labeling can have some impact on consumer behaviour and potentially on the calorie content of foods provided by restaurants. However, the impact has been relatively modest and not necessarily consistent across different contexts and studies.

In 2018, researchers for Cochcrane, an international research network, reviewed evidence to establish whether and by how much, nutritional labels on food or non-alcoholic drinks affect the amount of food or drink people choose, buy, eat, or drink. They considered studies in which the labels had to include information on the nutritional or calorie content of the food or drink.

The team combined results from three studies where calorie labels were added to menus or put next to food in restaurants, coffee shops, and cafeterias. For a typical lunch with an intake of 600 calories labelling may reduce the energy content of food purchased by about 8% (48 calories). The authors judged the studies to have potential flaws that could have biased the results.

Combining results from eight studies carried out in artificial or laboratory settings could not show with certainty whether adding labels would have an impact on calories consumed. However, when five studies with potential flaws in their methods were removed, the three remaining studies showed that such labels could reduce calories consumed by about 12% per meal. The team noted that there was still some uncertainty around this effect and that further research was needed to establish the size of the effect with more precision.

A meta-analysis published in the American Journal of Public Health in 2019 found that calorie labelling led to a small reduction in calories ordered and consumed. The impact was slightly larger in cafeteria or buffet settings compared to restaurant settings. However, the authors noted considerable variation between studies, suggesting that the impact of calorie labelling might depend on various factors such as the type of food establishment and the specific characteristics of the population.

Researchers in Australia in 2017 undertook a meta analysis of 186 studies on the effects of calorie labelling on consumers and an analysis of 46 studies on the impact on retailers. They concluded that displaying calorie information encouraged a reduction of 27 calories per meal for customers as a whole. For women, the reduction was 60 calories per meal meaning that it made very little difference to men. It also encouraged retailers to reduce the number of calories in each menu item by 15 calories.

The researchers admit that these are not particularly significant reductions but argue that the calorie savings could be significant over time for frequent diners.

Since 2018 restaurant chains with 20 or more locations in the USA have been required to label menus with calorie information. Fast food chain McDonald's voluntarily started labelling calories in 2012 which gave an opportunity for researchers to evaluate its impact. A long term study collecting receipts and administering a questionnaire to around 5,500 adults, adolescents and parents or guardian of school age children who visited McDonalds and a control group of outlets was undertaken in cities across New England between 2010 and 2014.

The research found that calorie labelling made very little difference to calories consumed. Adults consumed 80 calories fewer in McDonald's but adults in the control group of restaurants consumed fewer anyway without calorie labelling. In effect, the net difference due to labelling was 19 calories. It was slightly higher for adolescents but much weaker (13 calories) among school age children.

Staying in the USA but this time from the retailer perspective, an analysis of the menu items on sale at 59 large restaurant chains between 2012 and 2019 found that there was no change in the calorie content of menu items that had been offered continuously throughout the period. However, new items, introduced after the introduction of calorie labelling in 2018, had a mean of 113 fewer calories per item.

Why does calorie labelling seem to make very little difference to consumption? Part of the reason for this could be that people are not sure how a 1,000 calorie burger and fries relates to their overall recommended daily calorie intake. Survey data from Northern Ireland illustrates this point. Data from the Food Standards Agency’s Eating Well Choosing Better survey shows that only 20% of men and 21% of women could correctly state the recommended daily calorie intake for their gender (men 2,500, women 2,000). Some people thought it was as high as 3,500 for men and 3,000 for women. If a high calorie meal is thought not to make too much of a dent in your daily recommended intake then it is not surprising that some won't think twice about eating it.

While the evidence from around the world suggests that the reduction in calories consumed as a result of labelling is minor, the spur that it gives retailers to reduce calorie laden menu items could be more significant.

This may well be the most effective aspect of the policy in the UK.

How Did They Do That : UFC

Thu, 04 May 2023 16:19:55 GMT

Transferable learnings from brands that broke through.

Why do some businesses really take off despite the odds being stacked against them? In this series of articles we put the spotlight on companies and organisations who have broken through the hegemony to answer the question 'How did they do that'.

Each article will take a well-known brand and identify the things it did to become successful. We identify transferable lessons and highlight how research, insight and data analysis can help apply the lessons and strengthen business success.

We continue the series with UFC.

AI in market research: Is it just hype?

Fri, 21 Apr 2023 18:51:29 GMT

Artificial intelligence, when combined with human intelligence, gets results

Introduction

Artificial Intelligence (AI) has been making significant strides in recent years, with innovations and applications emerging across various industries. Advocates will tell you that AI will profoundly change the market research industry . Is this true or is it just a load of hype?

In this article we delve into the role of AI in market research, discussing its capabilities and limitations and how the industry should respond to these advancements.

AI's capabilities in market research

There are a number of ways in which can improve how market research is undertaken, analysed and interpreted. In summary:

�� Automating Data Collection and Analysis

AI can efficiently collect and process vast amounts of data from a wide variety of sources, such as social media, customer reviews or online forums. By automating data collection and analysis, AI can save researchers significant time and effort, allowing them to focus on more strategic tasks.

�� Predictive Analytics

One of AI's strengths lies in its ability to analyse historical data to identify patterns and trends. This feature allows market researchers to make informed predictions about future customer behaviour, market trends and potential opportunities, giving businesses a competitive edge.

�� Sentiment Analysis

AI-powered sentiment analysis tools can accurately gauge consumer emotions and opinions by analysing text data from various sources. This information helps market researchers understand how customers perceive a brand, product or service and make data-driven decisions to improve customer satisfaction.

�� Advanced Segmentation

AI can also help with advanced segmentation, identifying clusters of similar customers based on their behaviour, preferences and demographic information. This information allows businesses to tailor their marketing strategies and target specific groups more effectively.

AI's limitations in market research

Despite its many advantages, AI also has its limitations, which the market research industry must acknowledge and address.

�� Human Insights: AI can process vast amounts of data but it cannot replicate the human understanding of emotions, culture, and context that is crucial for market research. People's instincts, intuition and empathy remain indispensable for generating deep insights into consumer behaviour.

�� Bias & Ethics: AI models can sometimes perpetuate bias and unfairness, as they are trained on historical data that may include inherent prejudices. Ethical considerations are vital in market research and relying solely on AI could lead to skewed and harmful conclusions.

�� Adaptability: Market research often requires adjustments to methods and approaches, based on unique situations and contexts. AI models struggle to adapt as quickly and efficiently as humans, limiting their ability to deliver timely, accurate insights.

�� Black Box Problem: AI algorithms can be difficult to interpret, making it challenging to explain the reasoning behind their conclusions. In market research, transparency and clear communication are crucial for building trust and credibility with clients.

�� Implementation Barrier: Integrating AI into market research workflows can be costly and time-consuming, with many organisations lacking the resources to do so effectively. This creates a significant barrier to entry, limiting the potential impact of AI on the industry.

How the market research industry should respond

To embrace the benefits of AI while addressing its limitations, the market research industry should adopt a hybrid approach, combining AI-powered tools with human expertise. This approach will ensure that AI's capabilities are leveraged effectively while addressing its limitations through human insight and experience. This is the approach that we take at SPARK. We call it human made machine learning and it provides powerful business insight.

The industry should also prioritise ethical considerations and data privacy when using AI-driven tools. This includes being transparent about AI's role in research, ensuring data protection compliance and addressing potential biases in algorithms. Market research professionals must be equipped with the necessary skills and knowledge to work effectively alongside AI. This includes investing in training and education to help researchers understand AI's capabilities, limitations and ethical implications.

By adopting a hybrid approach the market research industry can harness the power of AI while addressing its challenges, ultimately driving innovation and growth.

Loneliness and young people

Thu, 13 Apr 2023 09:29:48 GMT

Why are one third of the under 30s lonely?

Loneliness is a common experience that can affect people of all ages, genders, and backgrounds. However, recent data from ONS shows that young people have been consistently more likely to experience loneliness than older adults.

The graphic shows that younger adults aged 16-29 are the most likely to feel lonely and that the incidence of loneliness decreases with age. Why is that?

In this article we explore the reasons why young people are more likely to be lonely than older adults. It will examine the impact of social support, social media, changing social norms, and mental health on loneliness in young people.

Percentage of adults reporting they feel lonely always, often, or some of the time by ageband

Source : Public opinions and social trends, Great Britain: personal well-being and loneliness, 22 March to 2 April 2023 , ONS. Values are the % at the start of each month.

The lack of social support

One of the primary factors that can contribute to loneliness is a lack of social support. This refers to the relationships and connections that people have with others, including friends, family members, and community members. Social support can provide individuals with a sense of belonging, emotional support, and practical assistance. However, young people may face challenges when it comes to building and maintaining social support networks. They may be more likely to move frequently for university, work, or other reasons, which can make it difficult to establish lasting relationships.

Additionally, young people may be more likely to experience social isolation due to their life circumstances. For example, they may be studying or working in a new city where they don't know anyone, or they may be in a long-distance relationship. These circumstances can make it challenging to build and maintain social connections, which can lead to feelings of loneliness.

The Impact of social media

Another factor that can contribute to loneliness in young people is the use of social media. While social media platforms can help young people connect with others, they can also contribute to feelings of loneliness and social isolation.

Research suggests that spending three hours or more on social media increases the risk of psychological issues such as mental health problems and loneliness. In addition, a negative experience on social media leads to an increase in the feeling of loneliness.

Social media use can create a sense of false intimacy and make it difficult to establish meaningful connections in real life. While young people may have hundreds of "friends" on social media, few of these will constitute close, meaningful relationships. This can lead to a sense of social isolation and disconnection. Additionally, social media use can contribute to feelings of anxiety and FOMO (fear of missing out). Young people often feel pressure to be constantly connected and available online, which can create a sense of anxiety and make it difficult to disconnect and recharge. This constant need to be online and connected can contribute to feelings of loneliness and social isolation.

Changing social norms

The way we socialise and connect with others has changed significantly in recent years, and young people may feel pressure to conform to new social norms. For example, social norms around dating, sex, and relationships have shifted, and young people may feel pressure to engage in casual relationships rather than forming long-term, committed partnerships.

These changing social norms can contribute to feelings of loneliness and social isolation. Young people may feel pressure to conform to new norms, even if they don't align with their values or preferences. This can lead to a sense of disconnection and loneliness.

For many generations the norm was leave school, possibly go to university and then get a job. But the economic downturn means that such a path is hard to follow for some which can add to the feeling of isolation because the interactions around work are not available. Additionally, many young people face pressure to succeed in their careers and may prioritise work over social connections. On top of this, many jobs now require some degree of remote working meaning that in-person contact is minimal. This can exacerbate any sense of isolation and loneliness.

Mental health

About one third of young adults experience mental health challenges, such as depression or anxiety, which can contribute to feelings of loneliness and social isolation.

Mental health challenges can make it difficult to establish and maintain social connections, as individuals may feel disconnected from others and struggle to engage in social activities. Additionally, mental health service are not always sufficient to meet local demand which can make it difficult for young people to reach out for support. This prolongs their exposure to mental health issues and resultant loneliness.

Conclusion

That one third of people aged under 30 feel lonely some or all the time seems shocking. Social media, home working, the economy and the fallout from the pandemic have contributed to a huge number of people feeling isolated and alone. In 2018, the government launched a strategy for tackling loneliness. Part of the strategy was to get a national conversation going to encourage people to reach out for support and be supported. The latest ONS data demonstrates clearly that more needs to be done.

Gunners still predicted to win Premier League. Probably.

Fri, 31 Mar 2023 22:21:44 GMT

Premier League title is still Arsenal's to lose

With the Premier League heading towards the final few games, we have rerun our predictive model to see who will win the title, which teams will qualify for Champions League football and which will be relegated.

The probabilities suggest that Arsenal will win the league from Manchester City. But it's very close. We also predict that Manchester United and Liverpool will join them next season in the Champions League.

At the bottom, we predict that Southampton, Nottingham Forest and Bournemouth will go down.

Our heatmap showing the probabilities for the position each team is likely to finish in.

We give the Gunners a 51% probability of winning with City on 49%. Last time we ran the model, at the end of the transfer window, Arsenal had a 55% chance of winning. The gap is narrowing. Squeaky bum time Gooners fans.

Manchester United has a 54% probability of finishing 3rd and has a 77% chance of qualifying for the Champions League along with Arsenal and Manchester City. Liverpool is predicted to finish 4th, slightly ahead of Newcastle and Spurs, but there's really very little to choose between all three sides. Liverpool historically finish seasons strongly which is why they are predicted to get the edge over Newcastle and Spurs.

There are only two other sides who have a chance of making the Champions League. Brighton has a 20% chance of a top 4 finish and Brentford has a 1% chance.

Premier League Probabilities Heatmap

At the bottom, the model predicts that Southampton (73% probability), Forest (60% probability) and Bournemouth (59% probability) will be relegated. There is still all to play for though as Wolves (20%), Leeds (17%) and West Ham (16%) are not out of danger yet.

Game on.

Undercooked

Thu, 23 Mar 2023 17:11:22 GMT

Is the bottom falling out of the meal delivery market?

Meal delivery companies, such as Deliveroo, Just Eat and Uber Eats have become a big part of everyday life. But is the bottom about to fall out of the market?

Delivering meals to the door is big business

Recent research by Food Standards Agenc y shows that over half of respondents in England, Wales and Northern Ireland (55%) have used an online meal delivery service with one in five of these (22%) doing so weekly.

The UK food delivery market (which includes restaurant and meal kit delivery) has experienced rapid growth over the past few years. The growth has been driven by changing consumer behaviour in the pandemic and technological advancements. Currently worth around £11bn per year and, according to Statista, it is expected to be worth approximately £12.5 billion by 2024.

The market is highly competitive with many people using more than one provider. A survey conducted by Savanta in 2021 found that 35% of respondents used two or more food delivery platforms. Our own research into meal kit providers shows that customers are willing to try multiple services to find the best fit for their needs. Switching between companies is common.

Can the remarkable growth of the online meal delivery companies carry on?

Consumer demand is coming under huge pressure as the costs of food, energy and housing increase. Recent ONS data shows that 43% have been cutting back on food and nearly two-thirds on non essentials as a result of the cost of living crisis. This is already having an impact on demand for food delivery services with a reported 12% decline in orders in January 2023 compared with January 2022.

Not only are the delivery firms being hit by consumer cuts, they are also being hit by increasing costs to their business - ingredients, fuel and energy prices have sky rocketed over the last year. Absorbing these will put a strain on profitability. Passing on the increases to the consumer could further dampen demand.

We can see the result of the fierce competition and difficult trading conditions by looking at recent performance. In March 2023, Just Eat reported a five-fold increase in worldwide losses in 2022 to €5.8bn (about £5.1bn). It has also sacked 1,700 of its delivery workers and reverted back to the 'gig' economy model that they ditched in 2020. Incidentally, they ditched the gig economy model because they said it was bad for workers so reverting back to it is unlikely to be popular with delivery operatives.

Deliveroo has also suffered in difficult trading conditions. In Q1 of 2022, delivery numbers were up by 19% but by Q4 they had fallen by 2%. Although the value of all transactions grew by 9% in 2022, this compares poorly with a 67% growth in 2021. Analysts believe that Deliveroo will make 9% of its staff redundant in order to control costs and is unlikely to see a profit before 2025.

Food delivery fights for survival

These are troubled times for the sector and a perfect storm of flagging consumer demand, lack of loyalty, increasing business costs and investors impatient for profit will make it a real fight for survival. Don't be surprised if there are consolidations in the market and casualties along the way.

The fight for survival might also speed up some of the changes in the market that we are already seeing. Things such as:

Increased automation so that food is cooked, packaged and delivered with minimal human involvement.
Greater use of technology to speed up logistics, distribution and route management.
The use of AI and machine learning to generate predictive menus displaying meals that an individual has the highest probability of buying. This will cut down waste, speed up fulfilment and optimise sales.

AI will also source the optimal provider or delivery hub based on current workloads, distance and previous performance. In the not too distant future your takeaway might be delivered by a drone from a fully automated factory producing and delivering meals 24/7.

The problem for the likes of Deliveroo, Just Eat and Uber Eats is that all of the above costs money and takes time. But time and money might very soon be off the menu.

Will ageism put a brake on over 50s returning to work?

Thu, 16 Mar 2023 16:14:04 GMT

Will the Chancellor's policies work or will workplace ageism stand in the way?

As part of the 2023 Spring Budget, the Chancellor announced initiatives to encourage over 50s back into work. These include an apprenticeship type programme for older people (referred to as 'returnerships') which will run alongside skills boot camps and sector-based work academies to give older people new skills to return to work. He also announced changes to pension allowances targeted at high earners and midlife MOTs helping people plan for retirement by doing a stock take of finances, health and skills.

Will these work or will other factors, including ageism, put a brake on the number who return?

Over half a million workers aged 50 or over have been lost to the job market since the pandemic. Latest data released by ONS shows that the number of over 50s classified as economically inactive (not employed, not looking for work and not in education) increased by 560,000 between October to December 2019 and October to December 2022. Of these, over 310,000 were aged 50-64, the pre retirement cohort most likely to be targeted by the Chancellor's drive to get older workers back into the workplace.

Why have so many over 50s become economically inactive?

The Over 50s Lifestyle Study (OLS), undertaken by ONS, sheds some light on why so many in this age group dropped out of the labour market. Retirement is a an important factor but so too is redundancy, illness and wanting a change of lifestyle.

Source : OLS graphic

Just under half of over 60s gave up work to retire with about one third of those aged 55-59 giving this reason. Redundancies and having to leave work due to illness, stress or mental health concerns are also important factors, particularly for the under 60s. There is also a significant number of people who left work because they wanted a lifestyle change and just didn't want to work anymore. Persuading these back into the job market could be tricky.

The OLS also reveals that the Chancellor's initiatives could land well with some in this age group. About 6 in 10 over 50s would consider returning to work, a figure which increases to 72% among those aged 50-59.

So what needs to happen to attract the over 50s back into the job market?

The OLS reveals that this age group would like greater support in relation to wellbeing and health but also flexible working, including working from home and the ability to fit work around other responsibilities, such as providing care. Upskilling, particularly around IT skills, is a requirement for some so the Chancellor's skills programme could be useful here. And of course, a decently paid job would also help.

Personal finances will be at the heart of many over 50s' decisions about whether to return to work. About half rely on a private pension so the relatively weak performance of private pension pots over the last year might have an impact here. While many people are confident that their financial provisions will see them through, people at the younger end of the scale tend not to be so - only 38% of 50-54 year olds are confident in their financial provisions compared with 55% of over 60s. The cost of living crisis might also tip more over 50s back into work.

There is, however, another issue, largely ignored in the current debate, that will need to be tackled to enable older workers to return to the job market. This is the issue of ageism.

A survey for Legal and General and the Centre for Economics and Business Research found that 52% of jobseekers over the age of 50 believed their age meant they were less likely to receive offers of employment. And it's not just their perception. A survey of managers for Chartered Management Institute found that three quarters (74%) were open, to a large extent, to hiring younger workers between the ages of 18 and 34. But that number dropped to just four in 10 managers (42%) when it came to hiring people aged between 50 and 64.

Ageism as a barrier to returning

Getting the over 50s back into work requires that ageism be tackled in all parts of the labour market from recruitment to doing the job.

Discrimination on the basis of age has been illegal in the UK since 2006 but attitudes against older workers can show up on an almost casual basis to perpetuate stereotypes; older workers are slow, ponderous and incapable of adapting to new ways of working. It's seen in perceptions that it's not worth sending older workers on training courses because they won't take it in or implement learnings. They are too set in their ways. They'll have to take time off because they are always ill. They are expensive and expendable especially in a time of economic crisis.

Ageism is at its worse when it's so nakedly done. Our research with older people showed the human impact of redundancy. People we spoke with weren't told outright that they were too old when applying for jobs; that would be illegal. Instead, hiring managers hid behind terms such as 'having the wrong cultural fit' or 'being over qualified' for the role.

We also see casual ageism in recruitment packs. We've seen job descriptions asking for applicants who share the values of a young, dynamic and energetic company. We've seen others that ask for digital natives - shorthand for millennials. How much corporate marketing, designed to position the company as a good employer, features older people? Not much, is the answer.

We see it also in the response to the current labour shortage in some industries. The debate in the various trade press is around filling the skills gap by attracting young people through apprenticeships or getting rid of the need to have a degree. There is nothing wrong with this and actually is a good thing but commentators are largely silent on the huge pool of older workers left behind. Similarly, workplace policies in relation to inclusivity and diversity can often ignore the needs of older people.

There is a huge pool of talent and experience wanting to come back into the labour market. Unless we address casual ageism and create a true culture of inclusivity, they will remain lost, despite the Chancellor's efforts.

OK Computer

Thu, 09 Mar 2023 12:39:18 GMT

We analyse user reviews of three well-known computer protection brands

The only way is down

Thu, 02 Mar 2023 15:41:14 GMT

House prices to fall by 10% in 2023

House prices have been immune to the economic pressures of Covid and the cost of living crisis. Up until now that is.

The latest data from Nationwide Building Society shows that house prices in the UK declined by 1.1% in February 2023 compared with February 2022. Prices have actually been falling for the last six months but last month was the first time that a year on year fall was recorded.

Other housing market statistics also point to a market in decline. The online sales portal Zoopla reports that 40% of its properties have seen a reduction in asking price, something almost unheard of in the last couple of years. Rightmove has reported that it took an average of 52 days for a home to come under offer in December 2022 compared with 32 days in May 2022. It's not surprising then that Land Registry data shows that sales in January 2023 were 7% lower than in January 2022.

So what does the future hold? Is this just a blip in the relentless march upwards or is the decline likely to be sustained? Looking at historical data can give us some clues.

The graphic below shows the average annual change in house prices. Values above the shaded area represent a growth in house prices. Anything in the shaded areas denotes a fall in prices. It shows that house price change has always gone up and done but usually the decline represents a slow down in growth rather than a fall in prices.

The graphic shows two occasions when house prices fell. The first was in the early 1990s when high interest rates and high unemployment made owning unaffordable for many first time buyers and existing owners. Consequently, the market was characterised by a slump in sales, negative equity, repossessions and three consecutive years of falling house prices.

Annual percentage change in house prices, 1982-2022, UK

The second house price fall was in 2008 during the 'credit crunch'. The 2008 global financial crisis was mainly caused by banks lending to people whose credit rating was poor and who subsequently defaulted on their repayments. Add to this higher commodity prices and rising levels of unemployment and the UK plunged into a recession lasting 5 consecutive quarters. The result was a lack of mortgage products and a crash of 15% in house prices. Interest rates, however, remained low and the crash lasted little more than one year.

Fast forward to today and we see some similarities with previous crashes. Similar to 2008, mortgage products are in short supply following the fallout from the Truss government 's mini budget in September 2022. Interest rates are increasing (similar to 1990 although at a much lower level) but unlike then, unemployment is comparatively low. This time, however, we have high inflation leading to real term cuts in income and a cost of living crisis. Buying a house is now more unaffordable and the uncertain economic outlook means people are either unable or unwilling to risk buying a new home.

History tells us that when prices start to fall they end up falling sharply. The decline in sales, lack of viewings and price reductions already working through the system suggests that this time will be no different. A decline in house prices in 2023 of 10% seems possible. If inflation comes down during the year and cheaper mortgages become more common, then we might see a smaller decline in house prices : the Office for Budget Responsibility is forecasting a 9% drop by Q3 of 2024.

The next two years are likely to be tough for the housing market and a far cry from the 'gold rush' of the last few years. For house prices, the only way is down.

People have been cutting back on heating and eating for over 1 year

Fri, 24 Feb 2023 16:30:06 GMT

New ONS data shows that cost of living crisis continues to have significant impact

The curse of the sneaky subscription

Fri, 10 Feb 2023 16:03:34 GMT

Why the sneaky subscription is bad for the consumer and bad for business

Many people have a subscription to services such as TV streaming, computer programmes, mobile phone apps and food, drink and beauty products delivered to the home. For the most part things work fine. The consumer gets what they need without the hassle of a break in service while the business gets a steady revenue stream.

But when the subscription is unwanted, the impact is felt by the unsuspecting consumer and, we argue, by the business too.

While most people are happy with their subscription others are very unhappy. There are two reasons for their unhappiness. The first is when it isn't clear that a free trial automatically converts into a paid subscription. The second is when a subscription automatically renews without people realising it or knowing how to cancel. It is these 'sneaky' subscriptions that are the real problem.

And it's a big problem. It is estimated that about one in four people have unwanted subscriptions. The Citizens Advice Bureau calculate that about half a billion pounds is wasted on these.

So why do businesses do it? The simple answer is that it's all about getting money in the bank. Last year, Business Insider ran a story suggesting that Amazon had known since 2017 that its user-interface misleads people into signing up for Prime. In response, a number of different fixes were trialled but all resulted in lower subscriptions and so were 'canned'.

With auto renewal, businesses calculate that the odds are stacked in their favour - there is more chance of someone not cancelling an auto renewing subscription than there is of getting someone to positively opt-in to the subscription.

And while this model means money in the bank, sneaky subscriptions could be storing up more trouble than they are worth. Research shows that 60% of people making online purchases read the contents of reviews on sites such as Trustpilot, Amazon, Google Play and Apple Store before buying. Have a read of user reviews on these sites and you will quickly see lots of examples of reviewers lambasting the company because they have been the victim of a sneaky subscription.

But it's not just the consumer who suffers. A pile of reviews that highlight sneaky subscription practices will negatively affect a business's overall rating and ultimately hit them where it hurts the most : the bottom line.

We do a lot of work analysing the content of Trustpilot reviews to guide clients on what people really think of their products and services. Two recent examples show how sneaky subscription can negatively impact user ratings. A business providing resources to trace family trees has a free trial period that auto converts into a paid subscription after a short time. A lot of people say that they were caught out by this because it wasn't clear when they signed up to the site. As a result, and despite users rating highly its service, it received a positive sentiment score for its cost of only 22% (the percentage of mentions about costs that were positive rather than negative) and a quarter of reviews were one star.

This example shows that an acclaimed service can suffer because of a sneaky subscription. When a sneaky subscription coincides with a poor service then the impact can be much worse. In the second example, a dating site's sneaky subscription was a contributory factor to 70% of its reviews being 1 star and one of the lowest positive sentiment with costs scores we've seen. Other factors contributing to the poor reviews include too many dormant profiles and unexpected fees when cancelling membership.

With so many people now consulting review sites before making purchase decisions, businesses have to be transparent about subscriptions and how to cancel an auto renewal. The Competitions and Markets Authority (the UK consumer protection regulator) has issued guidance to say that businesses should give consumers clear and prominent information about the subscription period, how much they will be charged, the length of the renewed contract period and how auto renewal works. The CMA suggests that a renewal reminder is sent out in good time. This is just guidance and so is not legally enforceable but it should be the minimum requirement.

A sneaky subscription might see the bank account grow but the damage it causes to a business's reputation isn't worth it in the long term.

Our view is that it isn't just the consumer who pays but ultimately the business too.

Cost effective insight

Tue, 07 Feb 2023 14:52:45 GMT

Three articles to help your insight budget go further.

With research budgets coming under pressure, the need for value for money insight is greater than ever. In these three articles we discuss how to maximise the insight from your research spend.

In the first article we show how your own and publicly available data sets can be analysed to provide purchase funnel type metrics.

The second article is about how to reduce research spend without compromising the insight you need. One of our suggestions is to ditch the brand tracker or at the very least strip it back to its core.

In the final article, we detail what a core set of questions for a stripped back brand tracker should cover.

Taken together, this series of articles will help you maximise insight while achieving much better value for money.

Feel free to contact us to discuss any of the points raised in these articles.

Gunners to win Premier League. Probably.

Wed, 01 Feb 2023 12:28:03 GMT

Premier League title is Arsenal's to lose

With the Premier League transfer window now closed thoughts turn to who will win the title, which teams will qualify for European football and which will be relegated.

We have developed a predictive model which helps us answer these questions.

The probabilities suggest that Arsenal will win the league from Manchester City. We give the Gunners a 56% probability of winning with City on 43%. These are decent odds in the Gunners' favour and the title really is theirs to lose.

We predict that Manchester United and Newcastle will join them next season in the Champions League. We predict that Southampton, Everton and Bournemouth will go down.

We have also drawn up a heatmap showing the probabilities for the position each team is likely to finish in. This shows that Manchester United has a 37% probability of finishing 3rd ahead of Newcastle. The Magpies are predicted to finish 4th but the model is close as to whether they finish 3rd, 4th or 5th.

Usual Champion League stalwarts Liverpool and Chelsea are not predicted to finish in the top 4. Of the two, Liverpool has the highest probability of qualifying at 39%. Chelsea is predicted to finish in 7th or 8th place. At this point it should be pointed out that the model does not take into account the impact of transfers in or out of the squad during January's window. We saw last year that Newcastle's spending helped improve their league position and with Chelsea having spent heavily in the window, the same could happen this year. Injuries, managerial changes and players coming in and out of form will also have an impact on where teams finish.

Premier League Probabilities Heatmap

At the bottom, although the model predicts Southampton, Everton and Bournemouth will be relegated, it will be very close. Nottingham Forest, Wolves, Leeds, Leicester and West Ham are all in the relegation mix.

We will update the model in April and publish the results then. Last season our model successfully predicted the winner, Champions League placings, Europa League and Europa Conference qualifiers plus the bottom 3 to be relegated. All in the correct order.

Back of the net.

The future of cycling is electric

Fri, 27 Jan 2023 17:54:58 GMT

E-bikes are on the rise and there's no back pedalling

Britain is not a cycling nation but all that could change if the e-bike revolution gets into gear.

E-bikes make up 9% of annual bike purchases in the UK which is around 165,000 sales. It's a low figure now but the potential for growth is enormous if we look at what is happening in other countries. In Germany and the Netherlands, sales of e-bikes are much stronger. In Germany they sell about 2m e-bikes a year and in the Netherlands 1m. About 40% of new bike sales in the Netherlands are e-bikes and about 5m e-bikes are in circulation .

Could this volume of use happen here? We think there are three good reasons why e-bikes will be the future for cycling in the UK:

Health benefits for the individual.
It's an environmentally friendly method of transport.
It can allow towns and cities to breathe again.

Health benefits

Electric bikes assist the user by given them some extra power as they pedal. And that's the key thing. You have to pedal the bike to get some assistance from the motor. E-bikes are not like motorbikes, they don't have throttles that power the bike without pedalling. Furthermore, the motor cuts out at 15 mph (25 km/h) and so after that you're on your own. The assist in effect flattens hills and is similar to cycling with the wind behind you. You have to put some effort in but you don't end up a sweaty, crumpled, mess at the end of your journey.

Research in the USA comparing riders of e-bikes with those of normal bikes found that e-bike riders’ heart rates reached, on average, 89 per cent of the mean achieved when riding a non-assisted bike . This gave them a moderate intensity workout. It concluded that e-bike users retained the majority of the cardiovascular health benefits associated with cycling.

Add this to the fact that people on e-bikes cycle further than people on normal bikes and you can see that e-bikes have a positive impact on fitness and wellbeing. If you can commute to work, run errands or visit places on your e-bike rather than by car, then that has to be good for the body and soul.

An environmentally friendly method of transport

Electric bikes are better for the environment. Recent estimates show that about 2.4kg of CO2 is emitted by cars for every litre of fuel (10.9kg of CO2 per gallon). The government has set a target for half of all short trips to be walked or cycled by 2030 in England. In 2019 more than 60% of trips between one and two miles were driven. If short journeys can be done by an e-bike rather than by car fossil fuel pollution will reduce. A report on the role of e-bikes in decarbonisation suggests that switching from a daily car journey to an e-bike can save an average 249g of CO2 for every 1km travelled.

This is really important because in the UK, air pollution, which car emissions contribute to, is the largest environmental risk to public health. The annual mortality of human-made air pollution in the UK is roughly equivalent to between 28,000 and 36,000 deaths every year. Anything that can decrease this number has got to be good.

Helping town and cities breathe again

The potential for e-bikes to breathe fresh life into our cities and towns is enormous. Britain is no longer a cycling nation and over the last 80 years m ost of our towns and cities have developed putting infrastructure for the motor car as a priority. As a result, we have ended up with urban areas suffering from congestion, pollution, safety risks from having to share space with cars and a negative impact on the well being of people.

Imagine being able to visit traffic free town centres and cities where only cyclists can go. Think of the environmental impact of less congestion and less CO2 and the safety aspect from not having to share roads with cars. No more traffic jams and no more hassle of trying to find somewhere to park. An environment that people would want to come to.

Such a scenario is not far away if, again, we look to the Netherlands and the town of Houten in Utrecht. Houten isn't completely car free but it's as near as you're ever likely to get. It was built about 50 years ago to give the priority to pedestrians and cyclists. There is a ring road circling the suburb which allows car access to the edges of each residential district. Internally, there are about 130km of cycle routes offering quick access to shops, schools and workplaces. Some cars are allowed in the centre but they are segregated and access restricted meaning that it is far quicker to cycle in.

It really is an eye-opener of what can be achieved by placing people before cars in the planning process.

Our friends electric

Obviously, you can't just impose Houten style segregation in already established urban areas. But there is more that can be done.

More and better cycling infrastructure for a start. Copenhagen, with wide and segregated cycling lanes, shows what can be achieved when people are prioritised in an existing city.

Here in the UK, planners should factor in e-bike usage when considering permissions for retail and residential developments and encourage more e-bike rental schemes to be established in every major town and city. Provision should be made for storing e-bikes at hubs such as railway stations, shopping malls and schools and charge points should be widely available.

The e-bike revolution is coming. We have a chance to make our towns and cities better. We will feel better for it as individuals. And we will feel better for it as a nation.

Combining survey findings with existing data

183:752711313 (Michael Wagstaff) — Thu, 19 Jan 2023 10:30:16 GMT

Integrating survey research, social media and CRM data to get complete business intelligence

Businesses rely on data to review performance and identify opportunities.

For many years the survey was the first thing marketers would reach for in their toolkit. What better way is there to gauge where you are on key purchase funnel metrics such as awareness, consideration, intention and action?

But good business data can come from others sources too and it's vital that brands are plugged into these. Business insight can be derived by applying data science techniques to existing data such as CRM or sales figures as well as publicly available datasets, such as social media and ratings & review sites.

Using survey data to set the baseline for measuring change

We find surveys really useful for assessing spontaneous and prompted awareness of a brand or product. This is a good baseline from which the impact of marketing and investment can be measured. Survey data is also really good at helping to understand a consumer's engagement not just with the product but also with the category that product belongs to and what they think about other brands in the competitive set.

We also use surveys to understand how a brand is perceived, how front of mind it is in buying (or using) situations and for identifying social, economic and environmental barriers to buying (or using) the brand.

We also ask about intention. Asking people how likely they are to buy your product or use your service is a useful indicator of future demand. Be aware though that there is often a wide disconnect between what people say they will do and what they actually do. Nowhere is this truer than in stated intention to buy a product or use a service.

Use internal business data to model and predict

A brand will have a lot of data internally that can be mined for insight. Sales, CRM and other internal data and even historic brand tracking data, can be used to undertake predictive modelling, market mapping, marketing ROI, revenue forecasting and to construct churn and retention ratios.

All this is vital in the day to day running of the business.

Use social media and review data to identify market position and optimise customer experience

Of great importance is the sway that ratings and review sites have on purchases. A huge number of people look online for product reviews as part of their purchase decision. The really interesting thing is that the number of five star reviews is not the critical factor here. Instead, it's what people say about the product - negative and positive - that really matters. This means that brands really need to pay attention to what is being said about them.

Ratings and review sites can be mined for customer experience insight and for deriving key drivers of brand growth through the identification of features and their importance. Text analysis can be used to identify the attributes most associated with a brand and its competitors and how distinct it is from its rivals in the mind of the consumer.

If a brand has a presence on social media, a network analysis can be constructed. A network analysis is used to understand the relationships between organisations by mapping the accounts followed by followers of the brand and then mapping the accounts followed in turn by their followers. By graphing the connections between them we can see which accounts are the most influential over the network. A segmentation can also be undertaken based on followers. The information provides insight on discreet groupings of followers and on the people, events and media content that interests them. This information will help with more effective targeting of existing and potential users of the brand.

Putting it altogether

Survey data will only tell you so much about your brand, product or attraction. Respondent recollection of their experience can often be hazy and lead to inaccurate results. CRM data will tell you how many customers you have, how often they buy and how much they spend. But it won't tell you the key drivers of their purchase decisions, the strengths and weaknesses of your product and how it compares with rivals and what your place is in the market

You need this insight to grow your market share. This is where deep analysis of social media and ratings & review sites are so important. Here, you can add the real voice of the consumer to your internal and survey data and get vital feedback from people who have used your product or service.

At SPARK, we integrate social media and ratings and review data with survey and CRM data to understand the market in which a brand operates and their place within it.

We start by understanding your business, its market and your intelligence requirements. We then identify the research approach needed to meet these requirements. For example, is there a need for brand health metrics, price optimisation, trade off analysis, identification of drivers of satisfaction, purchase or use or 'what if' modelling scenarios. We then assess what insight is already available from existing sources that will provide the required intelligence and which data analysis technique do we need to use to extract it. Once we know where the gaps in data are we will then commission a survey to fill them.

We call it Complete Business Intelligence and it provides the insight needed for great decision making.

Gym Resolution

Fri, 13 Jan 2023 16:45:24 GMT

Which gym is best for equipment and which one has the best classes?

The insights and data industry in 2023

183:752711313 (Michael Wagstaff) — Wed, 04 Jan 2023 11:09:58 GMT

What does the next 12 months look like for the insights and data industry?

Last year the research industry found its feet again after the pandemic. It was a year of DIY survey platforms, the need for speed, the increasing importance of data science and a year of acquisition and mergers.

What does this year hold? We look into our crystal ball to pick out some of the likely features of 2023.

A year of blending

Blending data from different sources will become increasingly important in 2023. By blending we mean combining analysis of existing data (sales and CRM, publicly available data from ratings & review sites and social media) with survey research. Blending achieves a much deeper view of business performance in a way survey research on its own never can.

The primary driver for this will be the realisation that businesses already sit on a huge amount of insight that can be readily extracted using data analytics and text analytics. Savvy businesses will only commission primary research to fill any gaps in existing data.

Cutbacks in the industry

With a likely prolonged recession and the resultant hit on consumer spending, 2023 will be a tough year for the insights industry. We are likely to see cutbacks in research budgets leading to fewer contracts, a squeeze on profits and ultimately job losses.

Of course, agencies will fight hard to retain clients by upping their client servicing game and promoting innovation but money talks and when revenue drops, jobs go. It would be no surprise to see a well-known agency go under and, if past trends are anything to go by, redundancies at senior level and recruitment freezes at junior levels.

Job cuts are, however, terribly counter productive because they lead to trouble further down the line. Expect to see a future deficit of experienced directors who understand a clients business and a shortfall of good SREs and RMs in a few years time.

The metaverse is this year's blockchain

Remember how industry sages told us how blockchain would revolutionise market research? This year those same sages are telling us it's all about the metaverse.

The metaverse is another one of those things that wise heads say will be important but few understand what it is and even fewer have any idea how to use it.

The simplest definition of the metaverse is that is a network of virtual worlds. People are immersed into the metaverse using virtual reality headsets or augmented reality technology. We are not saying that the idea doesn't have potential but in an industry which is very traditional in its outlook, the metaverse is not arriving in the market research world anytime soon.

Poor quality samples will be 2023's big headache

In the dash for speed quality has taken a hit. With so many survey providers having little or no control over who actually answers their questionnaires and, more importantly, whether people are actually who they say they are, it is little wonder that many surveys are unreliable. A client might be able to get hundreds of responses within a few hours of coming up with their questions but what is the point if the results are rubbish?

The market research industry really needs to act on bad samples. Survey fraud, inattentive respondents and 'Walter Mitty' panellists are more common than the industry cares (or dares) to admit.

Let's hope that in 2023 it addresses the problem.

Qualitative research will have a good 2023

Qualitative research will have a good year. It can be very effectively used to get the detail on why people behave and act the way they do. Behavioural research will continue to be a big topic in 2023 and qualitative research can untap a rich seam of behavioural insights.

It's not cheap but it represents value for the insight it provides. There are also a lot of seriously good freelancers and micro suppliers giving a first class service at an affordable price.

Argentina to win. Probably.

183:752711313 (Michael Wagstaff) — Sun, 18 Dec 2022 13:51:44 GMT

We predict Argentina to win but it's very close.

The Fifa World Cup comes to an end on Sunday and our final prediction is that Argentina will beat France but it will be close.

Over the course of the World Cup our predictive model, based on form and the capabilities of the squad, calculated the probability of each team progressing to the next round. How well did it do?

Here are the scores for each round:

It successfully predicted 12 out of the 16 countries to get out of the group stage. It didn't predict that Germany and Belgium would exit so early or that Morocco would go through along with Japan, South Korea and Australia.
It predicted seven of the eight teams progressing from the round of 16. It didn't see Morocco beating Spain but then ago who did?
The model didn't have a good quarter final only predicting two of the four winners. It's bittersweet that it correctly predicted France to beat England. In fact, before a ball had been kicked, it predicted an England exit at the quarter final stage.
It correctly predicted that France and Argentina would make it to the final.

At the start of the tournament, the model predicted that Brazil would win the World Cup but Croatia saw to that in the quarter final. Argentina and France were second and third favourites to make the final in our very first prediction.

Fast growing sports for children and young people

183:752711313 (Michael Wagstaff) — Wed, 14 Dec 2022 16:25:18 GMT

Football maintains popularity but handball is the one to watch

Costs of care

183:752711313 (Michael Wagstaff) — Tue, 13 Dec 2022 17:01:45 GMT

Care home residents spend a shocking £4.9b a year of their own money on care.

Driving Home for Christmas

183:752711313 (Michael Wagstaff) — Thu, 08 Dec 2022 11:52:15 GMT

We analyse the content of Trustpilot reviews to see how car breakdown services perform.

Earnings fall in real terms

183:752711313 (Michael Wagstaff) — Thu, 01 Dec 2022 11:43:31 GMT

Average weekly earnings decline by 2.6% once inflation taken into account

Qatar World Cup

Sun, 20 Nov 2022 15:09:24 GMT

We put Data Science up front to play the percentages for the FIFA World Cup 2022.

Who will win the World Cup? Will Mbappé inspire the French to retain the crown they won in 2018? Will Messi finally lead Argentina to glory or will Brazil's firepower up front sweep them to victory? How will England and Wales do?

To find out the answers to these questions you can spend hours poring over the form book, analysing previous results, goals scored and goals conceded. But you don't need to do that because we've done it for you. We've done the data science taking all these factors into account and predicting the result for every group match. From this we have worked out probabilities for who is likely to progress from the group stage into the knockouts by virtue of finishing first or second in each of the eight groups.

Probabilities for making it through the group stage

For each team we show the percentage probability of progressing to the knockout stage once our model has predicted the outcome of every match played.

Netherlands is the strong favourite to progress from Group A but it looks close between Senegal and Ecuador as to who will join them. We don't rate host Qatar's chances of making the round of 16.

We expect Argentina, England, France, Spain, Belgium, Brazil and Portugal to progress to the quarter finals alongside Netherlands. Germany, Croatia, Denmark and Uruguay have a high chance of going through in their groups but groups A,B, C and G look tight in relation to the second team going through to the knockouts.

The model is predicting England to go through from Group B with Wales and USA battling it out to be the second team to go through. At the moment the model predicts that the USA will just pip Wales.

Who's going all the way?

As things stand, before a game has been played, our model suggests that one of Brazil, Argentina, France and Spain will win the World Cup. Brazil is favourite with a 16% chance. England is on 9%.

The model predicts that England has a 57% chance of reaching the quarter finals but only a 31% chance of getting to the semis. It also calculates that Wales has a 14% chance of reaching the quarter finals.

If Costa Rica was to win the World Cup it was represent the shock of all shocks as the model predicts that there is virtually no chance of that happening.

After each round of matches we will feed the results into the model to update it and amend the probabilities. Check back in or follow us on Twitter for the very latest FIFA World Cup predictions.

The comparers compared

183:752711313 (Michael Wagstaff) — Fri, 28 Oct 2022 13:12:50 GMT

How do the price comparison websites compare?

Electric vehicles charge ahead

183:752711313 (Michael Wagstaff) — Tue, 25 Oct 2022 11:51:13 GMT

Headline inflation is only half the story

Thu, 20 Oct 2022 14:48:20 GMT

Essentials outstrip headline inflation rate

Retail sales continue to fall

183:752711313 (Michael Wagstaff) — Thu, 29 Sep 2022 11:22:22 GMT

Retail sales volumes (the number of goods sold) continue their downward path. Sales values are up but this is down to the impact of inflation on food prices.

Duty free affordability

183:752711313 (Michael Wagstaff) — Thu, 22 Sep 2022 11:26:30 GMT

Cutting stamp duty could make buying a home less affordable

The housing market is firmly in the government's sights as it grapples with the cost of living crisis. The Truss government believes that the key to solving the crisis is economic growth driven by lower taxation. For this reason stamp duty is in its crosshairs.

Stamp duty land tax (SDLT) is a tax paid on the purchase of a house or land in England and Northern Ireland. In Scotland, the Land and Buildings Transaction Tax (LBTT) applies and buyers in Wales pay the Land Transaction Tax (LTT). The amount paid depends on the value of the purchase and the application of different tax rates at different thresholds of value.

During the pandemic, a stamp duty holiday was introduced. In England and NI, the lowest threshold at which stamp duty was payable was raised from £125,000 to £500,000 between July 2020 and 30th June 2021 and set at £250,000 between 1st July and 31st October 2021. In Scotland, between July 2020 and March 2021, the lowest LBTT threshold was set at £250,000. In Wales, the lowest LTT threshold was raised to £250,000 between July 2020 and July 2021.

The idea was to kick start the housing market after sales had slowed down during lockdown. It seemed to do the trick because the housing market exploded back into life. Unfortunately, house prices also exploded as this chart using data from Nationwide Building Society shows.

Annual percentage change in house prices by month, UK

Source : Nationwide Building Society

In the year to June 2021, house prices in the UK grew 13.4% on average. Annual growth fell back slightly once the stamp duty holiday period ended before reaching another peak in March 2022 when average annual prices rose by over 14%.

The stamp duty holiday was not the only reason behind rapid house price increases - pent up demand, the post pandemic desire to move out of cities, low interest rates and supply constraints all had an impact - but stamp duty relief and rising house prices are highly correlated.

The impact on affordability, particularly on first time buyers, was considerable. The graphic below plots how much of first time buyer take home pay is swallowed up by mortgage payments. The period of the stamp duty holiday is highlighted. For each region affordability worsened during the stamp duty holiday.

First time buyer mortgage payments as a percentage of take home pay by Nationwide BS region, Q4 2019 to Q2 2022, UK

Source : Nationwide Building Society

In London, the amount FTBs had to spend on their mortgage increased to over 50% of take home pay. In Wales, it increased to just under a quarter (24.7%) and in Northern Ireland it increased to 24.1% up 3 percentage points. The increase in Scotland was less dramatic - the holiday ran for a shorter period of time - but the overall trend is still upwards since the end of the holiday period.

For most regions, the end of the stamp duty holiday offered very little relief with the percentage of take home income spent on a mortgage either remaining broadly the same or rising as interest rate increases started to bite.

So what is actually achieved by cutting stamp duty? Advocates say that it saves buyers some cash which in turn helps make a move more financially achievable - more money for a deposit for example. This helps first time buyers get on the ladder and enhances people's ability to trade up or to downsize. This frees up supply and more supply helps with affordability. On the other hand, the data suggests that it contributes to house price inflation which makes repayments less affordable.

The government will be hoping that it's the former rather than the latter that prevails.

Business Concerns

183:752711313 (Michael Wagstaff) — Tue, 13 Sep 2022 15:56:31 GMT

Inflation and energy dominate business concerns

And the winner is

183:752711313 (Michael Wagstaff) — Wed, 07 Sep 2022 11:23:44 GMT

Live action Disney remakes fail to impress fans

As Disney's live action remake of Pinocchio, staring Tom Hanks supported by CGI, is released we look at how other Disney live action remakes have performed.

Disney has remade 10 of its animated classics and in each and every case viewers have rated the animated original higher than its live action remake. Will Pinocchio buck the trend?

Renewable fuel

Tue, 30 Aug 2022 11:44:32 GMT

What it is, where it comes from and its impact on the environment

From faddy eater to big business

183:752711313 (Michael Wagstaff) — Thu, 25 Aug 2022 16:51:35 GMT

The once derided faddy eater is now big business

Almost one in four people in the UK avoid certain types of food because it causes them physical discomfort. The majority do so because they have an intolerance or are allergic to it. Nowadays, this is known as food hypersensitivity but go back a short while and people who avoided certain foods were thought of as faddy eaters.

Our understanding of how certain foodstuffs affect the body has moved on considerably in the last few years and as the labelling of potential allergens of foodstuffs in shops and supermarkets has become more routine, the labelling of people with food hypersensitivity in derogatory terms has become far less prevalent.

The numbers show why this is the case. The latest data from the Food Standards Agency's (FSA) Food and You survey, shows that 24% of adults experience adverse reactions to certain foods or avoid them in case they have an adverse reaction. The FSA data shows that women are more likely than men to be sensitive to certain foods (29% v 17% men).

Half of people with food hypersensitivity (12% of the whole population) have an intolerance to certain foods. Of these, 4 in 10 state cow's milk and products made from cow's milk is the cause of their food intolerance. Gluten is next on the list as the graphic below from the FSA shows.

Top 5 food groups most likely to cause an intolerance

Source : FSA, Food and You 2, Wave 4, August 2022

About 4% of the population have an allergy to certain foods with peanuts and fruit affecting one in four people with an allergy and molluscs and crustaceans affecting about one in six.

Given these numbers, it's no surprise that foods free from gluten, dairy and lactose is big business. Research by Kantar World Panel, published in July 2022, estimates that the free from market in the UK is now worth £3.17bn.

So, food hypersensitivity is worth big money at the supermarket but it has not really taken off in the meal kits and take-away delivery sectors. There are some meal kit providers catering for people requiring gluten free, lactose free, dairy free or low fodmap diets. But these meals tend to be part of a vegan or vegetarian range which does not necessarily meet all requirements.

Similarly, the likes of Just Eat and Deliveroo are missing a trick here by not ensuring that they cater sufficiently for the one quarter of the population who have a more specialist dietary need.

It really is important that businesses in these sectors have a better understanding of their markets. Perhaps they shouldn't be so 'faddy' in who they cater for.

Redundancies up as vacancies decline

183:752711313 (Michael Wagstaff) — Thu, 25 Aug 2022 15:13:09 GMT

Latest economic indicators make grim reading

The ONS publishes weekly data on key indicators of economic activity. We have looked at those relating to job vacancies, redundancies and company dissolutions, and added other ONS data where relevant, to see how things are looking. In short, it's not too good.

How do digital banks measure up?

183:752711313 (Michael Wagstaff) — Mon, 22 Aug 2022 11:13:56 GMT

Analysis of user reviews shows how Monzo, Revolut and Starling Bank perform

In this report we analyse Trustpilot reviews to see which of the digital banks comes out on top.

Will football coming home make any difference?

183:752711313 (Michael Wagstaff) — Mon, 01 Aug 2022 10:54:46 GMT

Encouraging more girls to play football in schools will be key

After 56 years of hurt football has finally come home. After the euphoria of England's 2-1 extra time win over Germany, the focus is now on the implications for the women's game. What will be the legacy of England's Euros win? Will it inspire a generation of girls to play football. From this girl can to this girl does.

Leaving a legacy of greater participation is a vital outcome for any major sporting event be it a football tournament or an Olympic games. But do major sporting events actually inspire greater participation in sport?

The answer is yes and no.

Weed et al undertook a comprehensive literature review of the participatory outcomes of Olympic games and major sports events to assess whether there was, what they call, a demonstration effect, whereby people are inspired to participate in sport by such events.

They applied the outcomes in participation to a behaviour change model (BCM) to assess where people are on a five stage pathway that goes from not being interested to being fully committed. The model is shown below.

The research found that there was some evidence (although it was weak) to suggest that there is a demonstration effect particularly in the run up to the sporting events. This effect was must likely to be seen through existing participants either playing sport more often or switching to new sports.

It was also found that existing sports participants responded positively in behaviour to big sporting events hosted in other countries. Again this link was quite weak but it was there nonetheless.

There was no evidence in the research that major sports events encouraged new participants. There was, however, some evidence that previously lapsed participants came back to sport.

In relation to the BCM, major sporting events have more of an impact on those already engaged with sports (Planning, Action and Maintenance stages) rather than at the more attitudinal Pre-Contemplation and Contemplation stages.

Their conclusion was that big sporting events do not inherently inspire new participants to sport. Things are, however, a bit more complicated than this. Programmes for developing individual sports have resulted in increased participation over time. Some of this is explained through people switching sports but it is also the case that a sport's exposure on a global stage combined with grassroots initiatives can start to change behaviours around new participation.

The implications for women's football in England

The research suggests that if the Euro win is to have a lasting legacy then the focus has to be at grassroots level with schools having a vital part to play - a lot of participation in sport is driven by being able to play at school either during school hours or at after school clubs.

Figures from Sport England's Active Lives survey shows that 20% of school aged girls played football once a week or more either at school or outside school hours during the 2020/21 academic year. This figure is up on the previous year (which was significantly lower due to coronavirus) but down on the 2018/19 academic year (pre pandemic). By way of comparison, the 2020/21 academic year figure for boys is 46%.

Participation in football for girls reaches its peak in the final two years of primary school (years 5 & 6) as the graphic from Sport England shows.

Percentage of school age girls who play football once a week or more by school year

Once at secondary school, participation drops off sharply so that by the time they reach years 9-11, only 15% of girls play. There are a lot of reasons that contribute to the drop off including academic commitments, lack of confidence and the fear of being judged . The lack of places for girls to play is also an issue. Schools need resources to provide opportunities for girls to carry on playing as they get older. This requires investment in staff, equipment and facilities. The big question is who will fund this: The FA? The government? Schools?

It's not just investment in schools that is needed. It's also in grassroots clubs. I coached a junior football team from when they were aged 5 through to 18. Up until age 11 the team was a mixture of girls and boys. After age 11, FA rules meant that the girls could no longer play alongside boys. Most of the boys carried on playing when they went to secondary school but all the girls stopped because there was nowhere for them to go. The club I coached at was typical of most junior football clubs - it was run on a shoestring by a small number of volunteers. There wasn't the resources or personnel available to set up a girl's team.

Unless there are more grassroots football clubs for girls and young women up and down the country then this pattern will carry on repeating itself. Again, substantial investment will be required to enable girls to carry on playing football as they get older. Again, who funds this?

The next generation of players like Alessia Russo and Nikita Parris are out there. They have been inspired by the lionesses and their fabulous win. But the win on its own will not be enough to encourage girls to play football - the research discussed earlier shows this - instead a huge investment in schools and in grassroots facilities will be required.

The challenge now is to make sure that this opportunity is not squandered.

Has data science killed the survey?

183:752711313 (Michael Wagstaff) — Wed, 27 Jul 2022 11:03:38 GMT

Analysing existing data can provide valuable insight on purchase funnel metrics

Yes, the survey has evolved over the years from face to face to telephone to online to restech platforms but asking people have they heard of a product, are they likely to buy it and what did they think about it afterwards is as old as market research itself.

That's changing now. Business insight can be derived by applying data science techniques to publicly available datasets. But is it any good?

Awareness and engagement with the brand, product or service

Social media sites such as Twitter and Instagram are really useful tools in measuring awareness about your brand and the depth of engagement with it. At a basic level, comparing the number of followers you have with the competitive set will tell you where you are in the market. Comparing your number of followers over time will indicate the success of your acquisition strategy.

For more in-depth insight you can map your network by plotting who your followers on Twitter and Instagram also follow. This map, known as network analysis, shows a brand who else is in their consideration set and the hierarchy of importance to consumers of the different brands.

You can also gauge awareness and interaction with your brand, service or product by using Google Trends. This tool is very useful because it enables you to see the relative volume of searches over a set time period. You can search for general terms about your brand and drill down to specific actions to see the strength of engagement consumers have with you.

To illustrate, we show trends in Google searches relating to the launch in the UK of TalkTV on 25th April and more specifically Piers Morgan's new interview show (Piers Morgan Uncensored) that was the centrepiece of the launch.

In the graphic, the blue line shows the volume of searches for 'TalkTV' between 22nd and 29th of April. Google records the volume of searches within a given time period as an index. The point with the highest number of searches is indexed to 100 with other values set on a scale of 1 to 100 in relation to this. So if there were half the number of searches on day 2 compared with day 1 then the score would be 50. We can see from the graphic that searches increased from 24th April reaching a peak just before the show was aired on launch day.

The red line and yellow line show volumes for two terms measuring interaction with the TV station and the programme. Searches for the term 'How to watch TalkTV' are shown in red and those for "how to watch Piers Morgan' in yellow. Volumes for both of these lines are shown relative to the blue line. At 7pm on 25th April, searches for TalkTV were at their highest volume during the week shown. Searches for 'How to watch Piers Morgan' were at 18% of the TalkTV volume at this time and searches for 'How to watch TalkTV' were 4% of the total.

From this graph we can see that awareness and interest in the station and the programme increased immediately before it launched and that up to about one fifth were sufficiently engaged to take action to watch it - or at least intended to do so.

We can also see that although searches declined in subsequent days, a series of mini peaks occurred with people searching just before the next daily broadcast of Piers Morgan's show. Interest and engagement with TalkTV peaked with the launch and is now running at a much lower level.

In fact the sharp fall off in searches reflects what has happened to viewing figures. Industry data (BARB) estimates that 317,000 people watched the first show but this had fallen to 62,000 a few days later. It seems, therefore, that there is a good relationship between intention to watch as measured by Google Trends and actually watching as measured by BARB.

Customer experience

Views on the customer experience can come from ratings and review sites. Survey data will only tell you so much about your brand, product or service. Respondent recollection of their experience can often be hazy and lead to inaccurate results. CRM data will tell you how many customers you have, how often they buy and how much they spend. But it won't tell you the key drivers of their purchase decisions, the strengths and weaknesses of your product and how it compares with rivals. You need this insight to grow your market share.

This is where consumer views on sites such as Amazon, Trust Pilot and Trip Advisor are so important. Here, you can hear the real voice of the consumer and get vital feedback from people who have used your product or visited your venue.

At SPARK, our text analytics capability combines machine learning with human created business rules to undertake topic and sentiment analysis to identify the factors that differentiate products and those that are key drivers of purchases. Applying this capability to ratings and review sites enables us to identify the metrics which consumers rate as important and then measure how a brand performs on these.

A recent example of how we use text analytics to assess the performance of three meal kit providers can be see here .

We also analyse followers of brands on Twitter and Instagram to map the most influential players in the market and to provide detailed insight into the likes, aspirations and motivations of followers that can be used for segmentation. We also use ratings and review sites to analyse the flows between brands - which brands people move from and who a brand loses customers to.

All of this is done without a survey.

So is the survey dead?

So, is the survey dead? No is the short answer. Survey data can offer greater d etail especially around awareness, can provide a link between consideration and action and can fill the gaps in existing data.

We use survey data for four key areas:

Awareness - we use both spontaneous and prompted awareness to assess depth of knowledge.
Attitudes and perceptions - what do people know about you and in what use cases do you come to mind?
Engagement - we measure how people engage with you, how frequently and why they engage and the touch points of engagement. We also measure their future intentions.
Customer experience - we measure satisfaction and advocacy.

We also use survey data to model the key drivers of engagement, assess customer satisfaction and undertake trade off analysis to give insight on the combination of features and pricing that will resonate best with your target audience.

Our view is that there is a lot of hugely valuable insight already out there and for many uses is every bit as good as doing bespoke primary research. It's also cheaper and quicker. But there are weaknesses around detail, availability for an individual brand and linkages and this is where survey data should be used in addition.

Markerters should utilise all sources available to them and not just restrict themselves to one tool. Expanding the toolkit to combine survey research with analysis of existing data will give brands complete business intelligence.

Don't forget your toothbrush

183:752711313 (Michael Wagstaff) — Thu, 07 Jul 2022 15:31:53 GMT

In recognition of Plastic Free July we have put together an infographic on the decomposition time of four everyday plastic items.

The stand out stat for us is that it takes an estimated 500 years for a toothbrush to decompose. Jaw dropping.

Leading the Whey

183:752711313 (Michael Wagstaff) — Wed, 22 Jun 2022 10:14:07 GMT

In this report we analyse thousands of Trustpilot reviews to provide insight on the performance of Bulk, Myprotein and Protein Works.

These are three of the best known providers of nutritional and fitness related supplements such as protein powders, shakes, bars and vitamin supplements .

We use topic modelling to identify the metrics that are important to reviewers and sentiment analysis to measure performance on these.

Happy Days

Wed, 15 Jun 2022 13:57:56 GMT

Four years ago we were the happiest we've been. Why was that?

With energy and fuel prices skyrocketing, double digit inflation, Ukraine and Brexit woes and Coronavirus still hanging around, it's no real surprise that we are still less happy now than we were prior to the pandemic.

The ONS Opinions and Lifestyle Survey asks respondents to state how happy they felt yesterday on a scale of 0 to 10, where 0 is “not at all” and 10 is “completely”. In the first week of April 2022, the happiness rating was 6.9 compared with 7.2 in February 2020.

If we look at our happiness over time, this time using the ONS Annual Population Survey (see graphic below), we can see that the happiest we've been in the last 10 years was in April to June 2018.

Happiness in the UK, April to June 2011 to July to September 2021

Source: Annual Population Survey, Office for National Statistics

What reasons were there to be cheerful back then?

Well, inflation was only 2.3% during this period and unleaded petrol was around 126p a litre.

This was the time of Prime Minister Theresa May and the never-ending saga of a Brexit deal. David Davies was Brexit Secretary with Boris Johnson Foreign Secretary. Both would resign a few weeks later. Jeremy Corbyn was between his 'oh, Jeremy Corbyn!' phase and his 'oh, Jeremy Corbyn' phase as Labour Party leader. He led them to substantial gains in the May 2018 council elections but would lead them to a shattering defeat in the general election 18 months later.

The 2018 World Cup began in June with England eventually reaching the semi-final and Harry Kane winning the golden boot for scoring most goals. The June high spot was a Kane hat trick in the 6-1 defeat of Panama.

Cambridge Analytica filed for bankruptcy having been found to have illegally misused millions of Facebook users' personal data for political purposes.

Gender pay gap reporting was introduced during this quarter in 2018. Research last year suggests that gender pay gap reporting is making a difference with a 19% decline in the difference in pay between men and women.

The weather was hot, hot, hot with it being the warmest May and June since records began. This warm weather made a welcome change following the Beast from the East cold snap earlier in the year.

And of course where would we be without a Royal wedding to bring a collective warm glow to the nation? May 2018 saw the marriage of Prince Harry to Meghan Markle. The story of Harry and Meghan has kept the likes of Piers Morgan and the red tops occupied virtually every day ever since.

So there it is, Brexit wrangling, political upheaval, footballing hopes ultimately dashed, hot weather, dodgy data and Harry and Meghan. Will we ever see those days again?

Cycling slips back

183:752711313 (Michael Wagstaff) — Fri, 10 Jun 2022 10:32:20 GMT

Data from the latest wave of Sport England's Active Lives survey shows that the number cycling is down on the previous 12 months

Sports on the up

183:752711313 (Michael Wagstaff) — Thu, 02 Jun 2022 11:47:14 GMT

Sport England data shows overall activity levels down but three sports on the up.

Measuring up online estate agents

183:752711313 (Michael Wagstaff) — Thu, 26 May 2022 11:30:13 GMT

How do Purplebricks, Strike and Yopa perform?

The use of online estate agents has increased recently as sellers look to save money. About 1 in 12 properties are now sold through an agent that exists exclusively online.

But are they any good and are the savings worth it?

In this research we have analysed thousands of Trustpilot reviews relating to the three biggest online estate agents to see what their customers really think.

Spike in insolvencies reveals 'cost of business' crisis

183:752711313 (Michael Wagstaff) — Tue, 17 May 2022 15:00:12 GMT

Business insolvencies higher than during pandemic as 'cost of business' crisis hits hard.

There is a lot of focus at the moment on the cost of living crisis and the impact it has on people. Running parallel to this is what might be termed as a 'cost of business' crisis with many businesses struggling to keep afloat in these difficult economic times.

Just how difficult things are for business can be seen from the latest government data on company insolvencies.

The number of businesses that failed in April 2022 was the second largest since the pandemic began. In fact, the 1,991 insolvencies will probably be revised upwards as a technical problem meant that a day's worth of data from April is missing.

Looking at the overall trend shows the scale of the issue. The graphic below shows that since February 2021, the number of insolvencies has increased sharply, driven by the increase in Creditors' Voluntary Liquidations. This is where shareholders of an insolvent company voluntarily elect to wind up the business. A liquidator is appointed to sell off the assets and distribute the proceeds to creditors. This is different from compulsory liquidation where it is the creditors who apply to liquidate the company.

The number of insolvencies in April 2022 is double that recorded in April 2021 and 39% higher than in April 2019.

Number of company insolvencies April 2019 to April 2022, England and Wales

Source: Insolvency Service using own data on compulsory liquidations and Companies House data on other insolvency types.

What has caused this spike in insolvencies and for how long will it continue? There are a number of factors at play here. The first is the huge increase in energy costs and the knock on effect on the price of materials and costs of transporting goods. In addition, the government's Covid recovery support schemes stopped in May 2021 and for some businesses having to pay back the loans that propped them up may have been a bridge too far.

Rising inflation and weakening consumer confidence has seen the demand for some consumer goods and services fall, leaving businesses in that sector exposed. Staff shortages will also have had an impact meaning that many businesses have struggled to produce things they can sell or provide revenue earning services. Finally, businesses that exported to the EU may have been tipped over the edge due to the post Brexit increase in regulations.

With energy prices and inflation both forecast to increase and consumer confidence weak as the cost of living crisis bites, the 'cost of business' crisis is not likely to go away anytime soon.

Final Premier League Prediction

Sat, 23 Apr 2022 10:18:35 GMT

With the Premier League season coming to a close, we again compute the probabilities of where each team will finish in the final table.

As we did before, we will be using our Dixon and Coles style football match predictive model to simulate thousands of possible permutations of results to predict the final standings probabilities.

The Premier League final standings probabilities

Note : The probabilities include matches played up to and including Thursday 21st April 2022

As the heat map shows, despite Liverpool's strong recent performances, Man City is still the favourite to win the league (62%). The top 4 looks to be Man City, Liverpool, Chelsea and one of Arsenal or Spurs. The model makes Spurs the favourite for fourth (55%) ahead of Arsenal (43%) with Man United's chances looking slim (2%).

Norwich and Watford look destined to go down, while Burnley (52%), Everton (35%) or Leeds (13%) could join them.

The model, and therefore these predictions, does not take into account the effect of managerial changes, injuries/return of key players or the importance of picking up points to the team (e.g. mid-table safe teams vs relegation threatened teams), all of which could have a big impact on a team's performance.

Whatever happens, the run in to the Premier League season looks set to be an exciting one. Probably.

House prices to fall this year

183:752711313 (Michael Wagstaff) — Wed, 13 Apr 2022 13:52:28 GMT

The warning signs are in the data

House prices have been about the only thing immune to the Covid-19 global pandemic. Prices have shot up due to huge demand from buyers looking for more space, a change in lifestyle favouring seaside and countryside living and the stamp duty holiday in 2021 saving buyers thousands of pounds. There is also no doubt that many households benefited financially from working at home during lockdown which meant more money to spend on housing.

Latest figures from the Nationwide Building Society's House Price Index show that the average house price in the UK is £265,300, an increase of 21% since the start of the pandemic. In the year to March 2022, prices increased nationally by over 14% as a lack of supply combined with strong demand added further upward pressure on house prices.

But how long can this go on? At what point do rising house prices become unsustainable and start to go into reverse? For the housing market to work efficiently it needs new entrants (first time buyers) coming into the market and existing owners trading up enabling those behind them to move up the property ladder. Once prices reach such a level that the flow of buyers through the market is impeded then serious problems arise.

There are plenty of warning signs that suggest prices will start to fall soon. The cost of living crisis and the recent rise in interest rates is obviously a big factor but the Nationwide data also reveals structural deficiencies in the market that may need to be rectified by a fall in prices : the increasing unaffordability of housing for first time buyers and the widening price gap between different property types.

In many regions owner occupation is unaffordable. The graphic below, from Nationwide Building Society, looks at initial mortgage payments for a prospective first time buyer relative to their take home (or net) pay.

First-time buyer mortgage payment as a percentage of net income

Nationwide Building Society graphic

For the UK as a whole, on average, first-time buyers need to spend over 30% of their take home pay on a mortgage. In London it is over 50%. In the counties immediately surrounding London (known as outer metropolitan) it is over 40%. In the South East, South West and East Anglia it is over 30% of net income. These are historically high rates of unaffordability compared with the long run average.

Add to this the spiralling costs of energy and fuel and with inflation currently at 7% (but likely to reach 10%) then owner occupation quickly becomes unaffordable for potential first-time buyers in large parts of the country.

The other structural fault line in the market is that it is becoming increasingly difficult for existing owners to trade up.

The price gap between properties is widening as the graphic below shows. The average price gap between a flat and a terraced house was £12,000 prior to the pandemic. Today it is £25,000. There is a similar widening of the gap between semis and detached houses with a smaller increase between terraced and semis. These are record high price gaps.

Price gaps between different property types

Nationwide Building Society graphic

The graphic shows that it is becoming increasingly more difficult for existing owners to trade up which in turn decreases the supply of cheaper properties. Those owners currently delighting in the soaring value of their home may have a nasty shock when they find they can't actually afford to move anywhere. Of course, they could always move to a cheaper region but not everyone has the ability to do this.

Normally a lack of supply means it's a seller's market and prices increase. This is what is being experienced at the moment. However, if that supply is too expensive for buyers, thereby weakening demand, something has to give to get the market moving again.

These structural flaws in the market can only be corrected by a realignment between house prices and income. This either means that income has to go up substantially or prices have to come down. My money is on the latter.

Getting the Measure of the Meal Kit providers

183:752711313 (Michael Wagstaff) — Thu, 31 Mar 2022 15:44:38 GMT

The meal kit market is big business. It was worth an estimated £1 billion in 2020 and is likely to increase to £1.5 billion by 2025.

In this report, we use text analytics to analyse the performance of three of the most well known providers - Gousto, HelloFresh and Mindful Chef - on key metrics discussed by Trustpilot reviewers.

The metrics were derived by identifying the most commonly raised topics in the reviews.

We then analysed thousands of Trustpilot reviews left in the six months up to the end of February 2022 and used sentiment analysis to score the providers on each metric.

Read our report to find out who comes out best.

World Poetry Day

Mon, 21 Mar 2022 15:48:03 GMT

Poetry is defined as an art form that uses the meaning, sound and rhythm of language to evoke an emotional response. To celebrate World Poetry Day, we look at the complete works of three of the most influential poets to identify the words that they most frequently use.

Below are portraits of Kipling, Stevenson and Poe derived by using some of the most common words (three or more letters) in their poetry. The size of each word represents its frequency of use. We've taken out words such as 'and', 'but' and 'the'.

Rudyard Kipling

Rudyard Kipling was born in India on 30 December 1865 and died on 18 January 1936. He is a renowned English poet and writer.

Using 523 poems by Kipling we have created a wordcloud portait.

His most commonly used words (three or more letters) are shall,man,men,one and old. The most common longer words (7 letters or more) are beneath, England, soldier, thousand and English.

Robert Louis Stevenson

Robert Louis Stevenson was born in Edinburgh, Scotland on 13 November 1850 and died on 3 December 1894. As well as being a Scottish poet he also wrote the novel Treasure Island.

Using 303 poems by Stevenson we have created a wordcloud portait. The most common 5 words used (three or more letters) are day, like, still, shall and love. The most common longer words (7 letters or more) are children, morning, flowers, friends and mountain.

Edgar Allan Poe

Edgar Allan Poe was born in Boston on 19 January 1809 and died on 7 October 1849, an American writer and poet.

Using 57 poems by Poe we have created a wordcloud portait.

The most common 5 words used (three or more letters) are thy, upon, bells, night and love. The most common longer words (7 letters or more) are flowers, nothing, thought, chamber and nevermore.

AI Generated Poetry

Google's Verse by Verse experimental AI can be used to generate poems in the style of various different historical poets. The AI uses two models, one that generates poetic lines in the style of the poet and another that attempts to understand which line best follows the previous line.

Below is a poem created in the style of Edgar Allan Poe by the AI. We composed the first line by using words commonly used in Poe's poems. The AI generated subsequent lines.

Thy beauty shall light the night
Filled with deep passion, from the wild air.
Laid her on these crystalline delight,
Leaving he r young flowers to its lair.

Enjoy World Poetry Day!

Premier league predictions update

Sat, 12 Mar 2022 08:22:08 GMT

What are the chances?

After around five weeks since our first analysis , we again compute the probabilities of where each team will finish in the Premier League. This additional 50 games or so allows our model to take into account the impact on form of new signings in the January transfer windows.

As we did before, we will be using our Dixon and Coles style football match predictive model to simulate thousands of possible permutations of results to predict the final standings probabilities.

The Premier League final standings probabilities

Note : The probabilities include matches played up to and including Thursday 10th March 2022

As the heat map shows, the title is still Man City’s to lose. Despite losing against Spurs our model gives them a 79% chance of winning the league this season. This probability is way ahead of Liverpool's at 21%. This is due to Liverpool having the more difficult run in: they play Man United, Spurs and Arsenal as well as Man City away from home.

The top 4 looks to be Man City, Liverpool, Chelsea and one of Arsenal, Man Utd and Spurs, with Arsenal being favourite at 57% to make the top 4 thanks to their great recent form.

Norwich and Watford look destined to go down, however, the final relegation spot looks less clear. The great form of Newcastle has seen them move out of danger and the poor form of Leeds, Everton and Brentford have pulled them into a relegation fight along with Burnley. Currently, Leeds (40%) are favourites to join Watford and Norwich in going down. Burnley (31%), however, have a high probability of being relegated and Everton (22%) and Brentford (12%) are also in trouble.

The six pointers

We are now into the last third of the season where matches seem more crucial as teams battle for European spots and against relegation. The most important of these matches are the 6-pointers - matches where teams battling for the same objectives face each other where it's possible for one teams to gain 3 points at the same time a rival is dropping 3. Given the importance of these matches, we focus now on what would constituent a good result for the teams involved.

Battle for the title

Liverpool has still to play Manchester City at the Etihad and if they are to have aspirations of winning the title, it is a game that they probably need to win. A win for Liverpool against their title rivals would almost quadruple their chances of winning the league compared with losing or drawing the game. If they draw or lose then their chances decrease by 19% and 81% respectively compared with if they were to win.

Top 4

Arsenal, Manchester United and Tottenham are the three main competitors for the top 4 and while Arsenal are favourites the final standings could be determined by a number of '6-pointers' with all three teams still to play each other.

If Manchester United are to get top 4 they may have to win both games against Arsenal and Spurs as even picking up 4 points may not be enough.

Tottenham have more acceptable options than Manchester United as even beating Arsenal but losing to Manchester United would give them a 29% chance of achieving top 4 .

Arsenal are in the strongest position. If they pick up at least 2 points from these two games they should be above 50% to get the top 4.

Relegation battle

It looks probable that Watford and Norwich are going down, with the final relegation spot between Brentford, Leeds, Burnley and Everton.

Out of the teams in conte ntion for the final relegation spot, Leeds's only 6-pointer is against Brentford. A win for Leeds in this game would increase their chances of survival by 50% compared with losing or drawing the game. A draw or a loss would decrease their chances by 5% and 32% resp ectively compared with not achieving that result.

Brentford have the most 6-pointers as they play Leeds, Burnley and Everton in the run in. Win all three and their chances of survival are almost 10 0%. However, even if they lose two of these matches as long as they win the other one, their chances of survival would still be over 80%.

Wins for Burnley against the Bees and the Toffees could see them safe. Given Burnley's knack of beating the drop in previous seasons, you wouldn't put it past them.

Wins for Frank Lampard's Everton against Burnley and Brentford should see them safe. Lose both and they are in trouble with a probability of being relegated around 50%.

The model, and therefore these predictions, does not take into account the effect of managerial changes, injuries to key players or government issued sanctions, all of which could have a big impact on a team's performance.

Whatever happens, the run in to the Premier League season l ooks set to be an exciting one. Probably.

Losing my religion

183:752711313 (Michael Wagstaff) — Mon, 28 Feb 2022 12:14:52 GMT

Are values more important than religious allegiance?

The 2021 census is likely to show a decline in the proportion of the population selecting Christianity as their religion. In the 2011 census, 59% identified as Christian which was down from 72% ten years earlier. It is thought that when results from the 2021 census are released later this year, the proportion reporting as Christian could be fewer than half.

In fact, the proportion of people who say they have no religion at all could well be over half the population.

Some indication of the move towards a more secular society can be seen through the British Social Attitudes survey (BSA). The BSA asks whether people regard themselves as belonging to a religion so is a different question to that used in the census ( What is your religion? ) but it gives a good indication of how things are moving. In 2011, 46% said that they did not belong to any religion. By 2018 that had increased to 52%.

The extent to which people search online for information about religious events is another useful indicator of trends. Care has to be taken with search data because Christian festivals such as Christmas and Easter are also major secular festivals and analysing trends for these terms is unlikely to provide much insight into religious allegiance.

With this in mind, we have looked at trends in search volumes for Lent (specifying the religious observance rather than the past tense of the verb to lend) and a specific part of Lent that signifies the end of the observance but is not a secular holiday: Maundy Thursday. We have also compared Lent with Ramadan (along with Eid which marks the end of Ramadan) and Diwali (along with Bhai Dooj which is celebrated during Diwali) to see how trends compare.

The graphics below use data from Google Trends for 2011 to 2021. Google Trends displays the volume of searches relative to the highest point in the timescale. A value of 100 represents the peak popularity of the search term. A value of 50 means that the search was half as popular at that time. We have added a linear trend line to each chart to highlight the direction of travel.

Volume of searches for key religious events, 2011 to 2021, Google Trends

The charts show that the volume of searches for Lent and Maundy Thursday have both experienced a similar decline over the 10 year period. Ramadan and Eid, on the other hand, show a sharp increase in the volume of searches throughout the 10 years. Searches for Diwali and Bhai Dooj also show an upward trend.

Based on this we can speculate that when the 2021 census results are released we will see a decline in the percentage who are Christian and an increase (and possibly a substantial one) in those whose religion is Islam (5% in 2011), Hindu (2% in the last census) or Sikh (less than 1% in 2011).

Of course, searching for information about a religious event does not necessarily mean that people belong to that religion. There could be all sorts of reasons why a search is undertaken (general interest, an educational reason, curiosity) and so we cannot apply trends in search terms to religious allegiance with any great precision.

That said, analysis of Google Trends data supports the findings of the BSA that formal allegiance to Christianity is on the decline in the UK.

This further brings up a question of whether it is still appropriate to ask people about religious allegiance in the census and in surveys. People might be turning their backs on formal Christianity but that does not mean that they are also turning away from the values that underpin it (compassion, love, truth, goodness). You only have to look at our response to world emergencies and the way the majority helped each other during the pandemic to note that our values haven't weakened. Would it be better to report values rather than religious allegiance?

Perhaps people are not losing 'religion' but expressing their beliefs differently.

Almost normal

Fri, 18 Feb 2022 16:43:28 GMT

Travel for retail and recreation purposes still below pre-pandemic levels nationally

With Plan B restrictions dropped in England at the end of January and similar easings in Northern Ireland, Wales and Scotland, to what extent are we back to normal?

Not quite is the answer as far as retail or recreation is concerned.

We've used Google Mobility data to plot the change in the percentage travelling to pubs, cafes, restaurants, libraries, theme parks, shopping centres, museums and cinemas on Saturday 12th February 2022 compared with the corresponding Saturday in February 2020 (pre-pandemic).

Across the UK as a whole, travel for retail or recreation is down 16 percentage points on pre-pandemic levels.

But as our map shows there are variations across the UK. On the map, the darker the shading the closer that district is to normality. The lighter the shading, the greater the gap from pre-pandemic levels of activity.

In parts of Wales, Glasgow, London and Leicester, visiting shops or leisure attractions is down by over 20 percentage points. On the other hand, travel to retail and recreation is at or above pre-pandemic levels in the Isle of Wight, Cardiff, Wokingham, Moray and Middlesbrough.

This time last year, travel for retail and recreation was 58 percentage points down on the previous year and so February 2022 feels almost normal compared with 12 months ago.

Almost but not quite.

City to win the league - probably

Fri, 11 Feb 2022 11:40:37 GMT

Rather belatedly, after several postponed games, all Premier League teams have now reached the halfway point of their season, having played at least 19 games.

This see ms like a good time to compute the probabilities of where each team will finish in the league.

The Premier League final standings probabilities

We have developed a final standings probability heat map based upon a Dixon and Cole s style football match predictive model. To generate the probabilities, we model the number of goals scored in a match as a Poisson distribution (the probability for a team to score one goal, two goals, three goals and so on in a match) and uses historical data to identify parameters to represent a team’s attacking strength, defensive strength and home advantage. Applying this model to the remaining Premier League fixtures, we can predict the final standings probabilities.

The Premier League final standings probabilities

Note : The probabilities include matches played up to and including Saturday 5th February 2022

The title is Man City’s to lose, as our model gives them a 94% chance of winning the league this season. Liverpool looks destined for second (81% chance) and Chelsea, despite some recent poor form, look like taking third (71% chance).

The heat map shows fourth spot to be one of the more competitive positions with Arsenal, Tottenham and Manchester United all with a reasonable chance of finishing in the top 4. As things stand, Spurs are slight favourites with a 38% probability of finishing in the top 4 compared with United’s 36%.

It is a similar story at the bottom of the table. Despite Norwich's recent good results, they look destined to go down along with Watford. The final relegation spot is most likely to be between Newcastle (57%) or Burnley (44%) but this could go down to the last game where Burnley play Newcastle. Home advantage could be crucial for the Clarets.

Norwich, Watford and Newcastle fans should not, however, resign themselves to playing in the Championship just yet. Similarly, Manchester United and Arsenal fans should not give up hopes of Champions League football as the model (along with all predictive models) cannot account for everything. In this case, the model is built using historic data to predict future results. It may not necessarily account for future events like a change of manager, players going in and out of form, injuries or Coronavirus absentees. It also does not factor in players who have joined or left the club in January and the effect that this has on team performance.

This leads on to a question of whether spending in the January transfer win dow makes much difference to where a team finishes in the league. Does spending big mean more points? We've had a look at net spend in the January transfer window over the last few seasons to work out the correlation between spending and points per game.

You can read this analysis here .

Is spending big in January worth it?

Tue, 08 Feb 2022 15:45:12 GMT

The January transfer window is often thought of as an opportunity for teams to cement their position or solve problems encountered in the first half of the season. There's a temptation to spend big to achieve that top 4 finish or to get yourself out of trouble at the bottom.

But does spending in the January transfer window make any difference to a team's results? In this article we analyse the data to find out.

Who spends in the January transfer window?

The heat map below shows the net spend of teams by their position in the table on January 1st for each of the last 10 seasons. We define net spend as expenditure on new players minus income received from selling players. A positive number means that the team spent more on buying players than they received from selling them. A negative number means that their income from selling players was greater than the amount spent buying them.

From the heat map it can be seen that much of the big spending in January is done by teams chasing titles, Champions League places or battling relegation with six of the top spenders of the season coming from the top 6 and three coming from the bottom 6. However, spending at the bottom is still more common with teams in the bottom 6 spending on average 3.5 times more than the teams at the top and 2.6 times more than teams in mid-table.

Net spend in January transfer window by Premier League position at 1st January, 2012/13 to 2020/21 seasons

This season, Newcastle has been taken over by the Public Investment Fund, the sovereign wealth fund of Saudi Arabia. A reported collective wealth of £320 billion makes the Magpies by far the richest club in Europe’s top five leagues. They have not been afraid to spend their money, with over 90 million net spend on transfers in their first window, the most spend in January by any team in the Premier League.

But is the big investment worth it? Does it guarantee survival in the Premier league? In this final section we investigate further the relationship between net spend and an improvement in results.

What difference does spending make?

Below are graphs from the previous nine seasons showing the correlation between net spend and the change in points per game (ppg) pre and post January transfer window - defined as ppg from the start of the season to the opening of the window compared with ppg from the closure of the window to the end of the season.

While several of the graphs show positive correlations, they are not strong and given that there are years that show negative or zero correlation, it is difficult to conclude that spending more in January guarantees an increase in the quality of results. An analysis using a team quality rating based on FIFA ratings showed similar results.

While this analysis shows insufficient evidence that higher January spending leads to better results on the pitch, it is not saying that all transfers in January are worthless.

From watching football, it is clear that individual transfers can have an impact on team performance. In the 19/20 season Manchester United signed Bruno Fernandes and saw an increase in ppg of 0.74. Last season, Jesse Lingard went to West Ham on loan in January and scored 9 goals in 16 appearances to increase their ppg by 0.33. Perceived player quality and transfer net spend are not enough to account for the short-term success of a transfer window. More intangible factors like settling in time, suitability to the team system and whether transfers solve a team issue are all important. Other factors such as injury or loss of form to key players and a change in manager and playing style can also affect the team’s performance.

Ultimately, the January transfer window is largely about solving problems but spending a large amount of money is no guarantee of success. But it's a risk many teams are willing to take.

Gender equality an uphill struggle at the Winter Olympics

Thu, 03 Feb 2022 17:16:00 GMT

Progress has been made but gender equality at the winter Olympics still has mountain to climb

The 2022 winter Olympic games in Beijing will play host to just under 3,000 competitors participating in over 100 events.

Under half of these competitors will be women. The International Olympic Committee (IOC) estimate that around 45% will be women. If this turns out to be correct then the 2022 winter Olympics will be the most gender balanced one yet. But it will still be some way short of parity and it doesn't feel right that a global event celebrating human sports achievement is still skewed towards male competitors.

Our graphic, based on IOC data, shows that much progress towards equality has been made in the last 50 or so years although much more remains to be done.

Percentage of female competitors at the winter Olympics, 1964-2018

At the 1964 winter Olympics in Innsbruck and the 1968 one in Grenoble, fewer than 1 in 5 competitors were women. It took thirty years for that figure to double to 36% (Nagano, Japan, 1998). By the time we get to Pyeonchang in 2018, we still only had 43% of competitors being women.

Why is the winter Olympics so gender imbalanced?

There are number of different reasons for the imbalance. Outdated attitudes to the role of women in sport - be there to crown the winner, rather than be crowned the winner - were endemic in winter Olympic sport (and more generally) throughout much of the 20th Century.

A lack of women's representation on the IOC until the early 1980s didn't help and some nations even today refuse to send women to the winter Olympics.

The IOC has been slow to introduce parity in the number of events competed for by women and men. Women's ice hockey was only introduced in 1998, women's bobsled in 2002 and ski jumping in Sochi in 2014. Unfortunately, there are still more events for men than there are for women in the winter Olympics programme (for example, there isn't a four person bobsled for women).

To be fair to the IOC, they recognise the issue and are taking steps to address the imbalance with member nations. As well as having more medal events at the games, they are introducing mixed gender events (there are four new ones in Beijing) and balancing the time given for men's and women's sports.

The question is are these steps enough?

The challenge for the IOC and member nations is to make sure that the 2026 Winter Olympics in Milan is the first to be gender balanced.

Is that asking too much?

Microsoft's game changer?

183:752711313 (Michael Wagstaff) — Thu, 20 Jan 2022 15:00:03 GMT

Acquisition of Activision Blizzard will see Microsoft as the third largest game maker

News that Microsoft is set to acquire Activision Blizzard for $70bn could have a huge impact on the home consoles market. The deal would make Microsoft the world's number 3 gaming company and could mean that a large number of games become exclusive to Microsoft's Xbox console.

Our recent analysis of thousands of user reviews shows that exclusivity of games is a major differentiator between the Xbox and the PS5 and Playstation wins this hands down, as our graphic shows.

We used text analytics to identify the important features for users of each console and measured their performance using sentiment analysis.

The PS5 has a 55 percentage point advantage on positive sentiment in relation to exclusivity of its games. If the acquisition goes through, the Xbox could turn that right around and then some.

It could be a real game changer.

They'll have to improve the controller first though!

You can read up on how we used text analytics to assess each games console by clicking here .

Plant Life :Veganuary 2022

Tue, 04 Jan 2022 14:43:36 GMT

With record sign-ups and big brand backing does Veganuary actually make a difference?

The annual Veganuary challenge encourages participants to follow a vegan diet for one month. In 2021, over 580,000 people signed up online to do the challenge - up from 400,000 in 2020.

Veganuary comes after many people, having spent the best part of a fortnight filling up on mince pies, chocolates and turkey, now resolve to live a healthier, more compassionate, lifestyle. But it also comes at a time when people are more concerned than ever about the environment. The recent COP26 on climate change showed how 14.5% of greenhouse gases are caused by food and dairy production.

So how ready are we to embrace plant life in a major way? The signs look encouraging.

Veganism has been one of the biggest food trends of recent times although it's difficult to say precisely how many people are vegan. Polls usually put the figure at around 3%-5% of the UK population.

The increase in popularity of veganism over the last few years is part of an overall trend to eat less meat. Surveys estimate that flexitarianism (eating a predominantly vegetarian diet but occasionally eating meat) accounts for about 14% of the population. A report on the future of food by Sainsbury's in 2019 predicted that by 2025 a quarter of British people will be vegetarian and half will identify as flexitarian.

Although veganism seems like a new thing it has actually been around for a long time. The Vegan Society was founded by UK animal rights advocate Donald Watson in 1944. Since then, the United Kingdom has retained a strong interest in the veganism movement. According to the Vegan Society, vegan milks, vegan cheese and vegan ready meals are purchased at a higher rate in the UK than anywhere else in Europe.

Motivations for doing Veganuary

Veganism started initially as a movement based on animal rights. Although this is still an important motivator, people go vegan for many different reasons.

Health and growing concerns about the environment are now common reasons. Data published by Veganuary shows that 46% signed up in 2021 due to concern for animals but for one in five, concern for the environment was their primary motivation.

Reasons for signing up to Veganuary

Source: Veganuary (www.veganuary.com)

Brands have invested heavily in veganism and Veganuary

It is not just individuals who have bought into the veganism movement: major food providers have invested heavily in vegan products. For Veganuary 2021, around 560 new vegan products were launched and 259 vegan menus were added by chain restaurants. All the major UK supermarkets now provide a vegan range. With increasing numbers of people looking for vegan food it is no surprise that brands are looking to get in on the action.

In previous years some brands, such as Greggs, released permanent items such as the vegan sausage roll and vegan steak bake, while others only provided plant based products for the month of Veganuary. This is now beginning to change. In January 2022, KFC announced that its vegan burger would become a permanent menu item and Burger King added vegan nuggets to its menu as part of its commitment to have half its menu meat free by 2030.

So brands are getting on board but is this due to corporate health and environmental concerns or is it all down to following the money? In one sense the answer is: does it actually matter? The key question really is whether Veganuary itself encourages people to take up veganism.

Does Veganuary make a difference?

The vision behind Veganuary is to have a vegan world with compassionate food choices, an end to animal farming, protection of the planet and improving health. The vision is really about changing human behaviour and the effectiveness of the initiative should be viewed in that context.

A behavioural change model often used in market research is the Transtheoretical Model (TTM) of behavioural change. The TTM, sets out a number of stages of change that take people along a pathway from having no intention to do something - in this case go vegan - to actually trying it (taking action) to maintaining the change. The stages as applied to going vegan can be seen in the graphic below.

Initiatives like Veganuary have an important role to play in moving people along the pathway. There are two points in particular where Veganuary could have an influence: (i) motivating people who are open to going vegan (Pre-contemplation) to find out more about the dietary choices available and (ii) providing support, encouragement and advice to those planning to go vegan so that they make the step to actually trying it (Action). This is also where the social media community surrounding Veganuary could also be helpful.

We posed the question earlier whether brands are jumping on the Veganuary bandwagon to make a quick buck. Big Food is motivated by sales so there is clearly an element of profiteering in this. But it could also be argued that there is a sort of symbiotic relationship between food producers and Veganuary. Vegan items on the menu in fast food outlets helps raise awareness of veganism and could help people make that jump from Planning to Action and beyond. At the same time food producers get more sales.

A key question is how likely are people to maintain a vegan lifestyle at the end of the month? A follow up survey of participants by the organisers of Veganuary found that 40% of those doing Veganuary 2021 intend to carry on being vegan. Even if this number is an overclaim (which is usually the case with polling on intended actions), it does still show that Veganuary plays a part in helping people change to a healthier, more sustainable and animal friendly lifestyle.

In conclusion, veganism is a dietary choice which is increasing in popularity. Veganuary helps to promote the vegan diet. For some, Veganuary is just a month-long challenge, for others it is to test a potential new diet and the first step in a long-term behavioural change.

With more people following a vegan diet and brands producing new ranges to meet growing demand, Veganuary has become an established event in the calander.

Its challenge is to be more than just an annual try out for a month.

Lonely this Christmas

183:752711313 (Michael Wagstaff) — Sun, 19 Dec 2021 19:49:03 GMT

The young rather than the old are most likely to be lonely this Christmas

Picture in your mind an image of a lonely person. Chances are you'll picture an elderly person who lives alone and hasn't seen anyone for weeks.

While this stereotypical lonely older person exists, it's actually younger people who are most likely to feel lonely.

Data from the latest wave of the ONS survey Coronavirus and the social impacts on Great Britain show that one in four Britons feel lonely some or all of the time. For people aged 16-29, that figure is one in three. For people aged 70 and it's just under one in six.

A huge number of young people live with some degree of loneliness in their lives. Only 11% of 16-29 year olds say they have never felt lonely.

And it's worse for women. Women are much more likely than men to feel lonely - 29% v 18% across all age groups. ONS doesn't break down the stats by age and gender but logic suggests that younger women are more likely than younger men to feel lonely.

Why do so many young people feel lonely? The restrictions due to the pandemic are obviously at play here. If you've not been able to socialise you are bound to feel lonely. That said, surveys prior to the pandemic also pointed to higher levels of loneliness among young people. Part of this is related to where they are in the lifestage. Young people are more likely to have just left established networks to go to university or start a new job in a different town which can amplify the feeling of loneliness.

Some commentators point the finger at social media. Facebook, Instagram, Twitter, Snapchat and TikTok are great at connecting people from around the world and there is no doubt that they can be a force for good in helping combat loneliness. At the same time, social media can be toxic, especially for women and girls, and to be trolled online can have a profound impact on how lonely and isolated people feel.

The shallow nature of social media friendships can affect loneliness too. Psychologist Robert Weiss makes an interesting distinction between social loneliness and emotional loneliness. Social loneliness is a lack of contact with others while emotional loneliness is when your contacts do not provide support, affirm your identity or create a sense of belonging. This means that it's possible to be lonely even if you have lots of social media connections.

This leads to a wider point that being alone and being lonely are not the same things. You can be lonely while living with other people - such as a household full of relatives at Christmas - or live alone and not feel lonely.

For most people Christmas is a time of happiness. But for many others it can be incredibly lonely. The stereotypical image of the lonely older person can mask just how endemic loneliness is among the young.

#BeKind.

Final prediction

Sat, 18 Dec 2021 20:33:08 GMT

Each week we rank Strictly contestants based on social media analysis. We call this the Strictly Sentiment score.

During every show we collect tweets on each contestant and extract the sentiment from them using a model that we built especially for Strictly. We use the sentiment score to rank the popularity of each contestant.

We then combine our Strictly Sentiment analysis with the judges scores to predict who will finish in the bottom two and go through to the dreaded dance off.

Strictly Sentiment leaderboard

It's the final!

If the winner was decided by Twitter sentiment then Rose wins it.

S trictly Sentiment score

The Strictly Sentiment score is derived from a sentiment analysis of tweets. Using natural language programming and machine learning we classify each tweet as positive, neutral or negative. Tweets are filtered so that they only relate to a celebrity's appearance on the show rather than what they do in their 'day job'. Only tweets made during the show and up to the time that the public voting phone lines close are included. This is to ensure that tweets relate to Strictly.

We then add up all the positive tweets for each contestant and assign a score between 1 and 100. This score is based on the relative distribution of positive tweets. We do this to make it easier to compare and contrast Strictly Sentiment scores.

Mutual Attraction

183:752711313 (Michael Wagstaff) — Fri, 03 Dec 2021 13:01:13 GMT

Building mental availability to attract visitors

For visitor attractions giving people a great experience is key to getting them to return.

But how do you get people there in the first place? And how do you stand out when there are so many local and national attractions battling for attention?

This is where building mental availability can help. Read our new report to find out how.

Text analytics in action

Wed, 17 Nov 2021 16:48:40 GMT

We apply text analytics to user reviews to see who is winning the console wars

There is so much information out there about brand, products and people. Ratings and review sites, social media and customer feedback channels have masses of data that can give you important access to the voice of the customer.

Most of this incredible insight is unstructured and this is where we help. We make sense of huge amounts of free form text by categorising it into coherent topics and extracting sentiment from it. Text analytics can help you understand what your customers care about, what they feel about your products and how they engage with them. It can help evaluate how products compare and their relative strengths and weaknesses.

Further,our text analytics identifies the drivers of satisfaction, improves customer retention and increases profitability.

In this new report, we analyse thousands of user reviews on the PS5, Xbox and Nintendo Switch to demonstrate how we undertake text analytics.

COP OUT

Sun, 31 Oct 2021 15:18:09 GMT

Are we still concerned about the environment and climate change?

Over the last 18 months the pandemic has been the number 1 concern for most people. Now, with the start of the UN Climate Change Conference (COP26),the environment is back in people's minds.

The recent IPCC report found that the Paris Agreement goal of limiting global warming to 1.5 degrees C will be exceeded by 2040. This has led to the re-emergence of environmental groups and figureheads promoting activism and a greater discussion about the environment.

Now that we are beginning to emerge out of the pandemic, the focus once more is on the role of individuals in addressing environmental concerns such as climate change. By analysing trends in energy consumption, travel, food sustainability, recycling and environmental activism, we ask whether people are doing enough and what more needs to be done.

The report also discusses whether an individual can pick and choose how they play their part in addressing environmental concerns, whether it's ok to be an environmental hypocrite and how an over reliance on s elf-motivation coupled with insufficient governance is hampering the fight against climate change.

Participation in sport by people with a disability badly affected by pandemic

183:752711313 (Michael Wagstaff) — Mon, 23 Aug 2021 17:46:24 GMT

The challenge for sports bodies in England is to get participation back up to pre-pandemic levels and then devise strategies for growth that can be sustained.

A network analysis of comedians who played the Edinburgh Fringe

Thu, 19 Aug 2021 15:33:57 GMT

Or why Dara O'Briain is the centre of the comedy universe

The Edinburgh Fringe is one of the world's largest arts festivals, open to anyone with any type of performance allowed. It is, however, best known for its comedy.

To celebrate the 2021 Fringe festival, we looked at stand-ups from the previous Fringe to see how they are connected to one another through their followers on Twitter.

To do this we used a technique called network analysis. Put simply, network analysis is the graphical representation of the relationships between variables (known as nodes). In this case our nodes are comedians who performed at the last Edinburgh Fringe but nodes could be brands, services, people, words or anything where you're interested in how it relates to something else.

For our Fringe analysis we used network analysis to identify clusters of comedians who share the same followers, the comparative size of their networks, which comedians are important in driving followers to others and who are the most influential in their networks.

We consulted the British Comedy Guide to create a list of comedians who performed at the 2019 Fringe festival and then collected their Twitter followers.

We then created a network between comedians where each node represents a comedian and the node size represents the number of connections to other comedians. You can see this in the graph below.

In the graph, connecting lines, known as edges, represent comedians sharing at least 50% of followers with a minimum threshold of 10,000 shared followers.

It is known that people who share connections on social media tend to form clusters or communities. The network analysis shows that there are four main communities with each one defined by a different colour. Within each there is a larger node connected with smaller nodes. If you hover your mouse over the nodes it will display the name of the comedian and the number of degrees (or links) with other comedians.

Dara O'Briain has the most connections and therefore the largest node. Dara also performs very highly on what are known as network centrality measures. These are important because they identify the biggest influencers in the network. O'Briain has the highest degree centrality measure. This is a measure of the number of connections a node has. Important nodes have more connections and none are more important than Dara O'Briain.

It is not just how many connections a node has that's important, it's also the importance of the nodes that it is connected to. A value known as Eigenvector centrality measures this. It decides the importance of a node by measuring how many other important nodes it is connected to.

The third measure we look at is known as betweeness centrality. This quantifies how many times a particular node comes in the shortest chosen path between two other nodes. The nodes with high betweenness centrality play a significant role in the communication and information flow within the network and can have a strategic control and influence over others. A good illustration of this is Lucy Porter in the yellow community (which we label Millican Izzard). The shortest path between Eddy Izzard and Sarah Millican goes through Lucy Porter so although she does not have the most connections, she has a strong influence in the network.

Let's now have a look at the four communities in more detail. We look at who is in them and what connects them, who has the most connections (known as degrees) in the group and who has the most influence as measured by their betweeness centrality and Eigenvector centrality score.

We also look at what differentiates each community by identifying the top 5 Twitter users that are unique to that community. These are Twitter users that appear in this community but not in any of the other communities.

The largest community centres around Dara O'Briain. Present in every edge and connected to every other person in the community and many outside of it, Dara O'Briain is central to this Edinburgh Fringe based network. This is largely down to him being the host of the long running panel show Mock The Week, where he comes into contact with a lot of comedians. This community also contains a number of Irish comedic links - Neil Delamere, Alison Spittle, Andrew Maxwell, Eleanor Tiernan, Ed Byrne and Jason Byrne.

What differentiates this cluster from the others? The top five unique users followed on Twitter by those involved in this community are Brian O’Driscoll (former Irish rugby union player), RTÉ News (Irish and International News) , Des Bishop (Irish-American stand-up comedian), thejournal.ie (online Irish newspape r ) and Amy Huberman (Irish actress and writer).

The presence of Irish celebrities is not surprising given the large number of Irish comedians within the community. It is not until you get down to the 14th most unique user, Milton Jones, that you get a suggestion of a link with Mock The Week. That Mock The Week is not a differentiator is to be expected. It is a show closely connected to a large number of comedians and has a large number of followers in all the other communities.

It is also notable that there are two news sources in the top five most unique users, which suggests that Irish news and current affairs has an important differential for this community.

The second biggest community is shown in blue. Present in all but one edge, Nish Kumar is central to the community and linked with the most people in it. In fact, many people in the group can be linked to Nish Kumar through appearing on shows that he has been a big part of: Ivo Graham, Sarah Kendall, Phil Wang, Desiree Burch and Sophie Duker have appeared on the News Quiz, a radio programme that Kumar hosted. Ed Gamble, Lou Sanders, Sarah Kendall and Phil Wang have been long term contestants on Taskmaster alongside Kumar and Desiree Burch has been on The Mash Report, a satirical comedy news series hosted by him. Josh Widdicombe links the rest of the comedians in the community. He was a regular on Stand Up For The Week with Seann Walsh, starred in Hypothetical and The Last Leg in which Maisie Adams has appeared and has been a host on Fighting Talk, a radio programme that Ivo Graham has featured on.

In relation to differentiation, the top five unique users followed by members of this community are Lolly Adefope (comedian), Michaela Coel (actor), Akala (rapper), Sarah Pascoe (comedian) and Owen Jones (political commentator and columnist).

Taskmaster contestants Lolly Adefope and Sara Pascoe provide evidence that the show may be an important influence in this community, while Owen Jones suggest that left of centre politics may also be important. Akala has also been vocal about political issues on several occasions. He was a big supporter of Jeremy Corbyn and signed a letter (along with other musicians) declaring support for Corbyn in the 2019 election. The top ten most unique users also includes Sadiq Khan, Jeremy Corbyn, Alexandria Ocasio-Cortez and Ed Miliband. This political influence could be related to tv show The Mash Report which has in the past aired segments critical of Brexit and the alt-right.

The third community, shown in yellow, contains two main nodes: Sarah Millican and Eddie Izzard. In this community, members are linked by appearing on the same shows. Members have made appearances on Live at the Apollo, with each person apart from Tony Law, appearing in the show at least once. Sarah Millican, Lucy Porter and Zoe Lyons were featured in a Live at the Apollo special, All Girls Live At The Apollo, highlighting women in comedy with Zoe Lyons also featuring in Pride Specials with Eddie Izzard hosting.

The top five unique users followed by members of this community are Sarah Brown (Chair of charity Theirworld and Executive Chair of Global Business Coalition for Education), 10 Downing Street (Office of the Prime Minister), Sarah Silverman (comedian), Sue Perkins (comedian and presenter) and Jennifer Saunders (actor, comedian and screenwriter).

Sarah Millican has appeared on Sarah Brown's podcast, 'Better Angels with Sarah Brown' and alongside ex PM Gordon Brown. Sarah has also met with Eddie Izzard in Downing Street after Izzard completed 43 marathons for Sport Relief. The other three in the top five most unique users are all female comedian or comic actors which follows in terms of the All Girls Live At The Apollo link. The 7th most unique user in this community is Matt Lucas who, like Sue Perkins, has been a presenter on the Great British Bake Off.

The last of the four main communities, shown in green, is centred around Jon Richardson. He has been in shows heavily featuring the other two people in the group. He starred in Meet the Richardsons with his wife, Lucy Beaumont and captained his 8 out of 10 Cats Does Countdown team where Nick Helm has been a guest several times.

The top five unique users followed by members of this community are Danny Dyer (actor), Ant and Dec (presenters), Joe Wilkinson (comedian), Keith Lemon (comedian, actor, presenter) and Michelle Keegan (actor). While Joe Wilkinson has been a regularly appearing as Rachel Riley 's assistant on 8 out of 10 Cats Does Countdown, the others are from various different shows and soaps. This suggests that the community is differentiated by its love of soaps, scripted reality shows and entertainment.

What does this all mean?

At its simplist level, if you're an aspiring stand up playing the Fringe then get connected with Dara O'Briain!

There are, however, other uses that the results of the network analysis can help with.

Hosting - Ho sts for shows could be chosen by looking at the centrality of a comedian either within a community or the entire network, depending on the specifics of the show.

Collaborations - It could be used to work out which comedians could collaborate with each other. Those that share a link are likely to share a similar audience.

Growing an audience - Comedians could associate with others with whom they don't share an edge (link). The benefit of this would be that both comedians could cross migrate each others followers.

Scheduling events - The Edinburgh fringe could look at a network analysis to decided where to place individual comedians e.g. by having comedians from the same community performing in succession in a similar area. A similar process could be looked at for scheduling TV shows with the aim of retaining and building on viewers.

Material for shows - By looking at the top user accounts that followers of individual edges (links) follow, themes could be identified for new performance material.

Business applications of social network analysis

In this article we have applied network analysis to comedians but it can be applied to marketing and business situations to provide insight.

In the example above the nodes represent comedians but they could just as easily represent a brand, a product, an organisation or a service. In an international network analysis, each node could be a country.

Edges, or the links that connects nodes, represent relationships between them. In the example above they represent Twitter connections but they could also represent buyers of a product, users of a service and strength of feeling in relation to brand attributes.

Twitter is a great source of data for network analysis but it is not the only one. Survey data, a CRM database and even large volumes of unstructured text can be utilised. Network analysis does not have to be quantitative, it can be done qualitatively too.

Network analysis can be used to understand how the market operates by graphing the network of brands or products in the competitive set. We can calculate the centrality measures to assess things like a brand’s power, activity and communications effectiveness. This will identify what a brand needs to focus on in order to grow.

Network analysis can be used to generate a segmentation. The analysis will identify clusters of people within the same consumer or social circles. Often this can be related to stated needs. From this you can identify the cluster with the most influence or those most likely to spend and target your marketing accordingly.

It can be used to identify marketing messages that resonate by undertaking a keyword analysis. In a keyword analysis, first a network map is created to understand how the various keywords are linked (based on their associated connotations) to a brand. Based on this mapping, various messages can be created to suit the needs.

Network analysis is really good for getting a detailed understanding of service use, needs and gaps. Our researchers used network analysis to qualitatively map formal and informal support providers for people with a life limiting medical condition. It identified the hierarchy of support services required and where investment was needed to provide new services.

For more information on how we can apply network analysis to help your business grow, please contact us.

The gold standard

183:752711313 (Michael Wagstaff) — Thu, 29 Jul 2021 17:38:27 GMT

Do the Olympic Games boost participation in sport and what can sports in the UK do to build on the increased exposure?

The Olympic Games are an expensive business but are they worth it? Does the host nation win more medals and is there a growth in participation at grassroots level?

The Tokyo Olympics are estimated to have cost in the region of $15.5b to host. This cost relates to investment in new stadia and facilities, renovation of existing ones and investment in infrastructure throughout the city.

The estimated costs for Tokyo 2020 are in the same region as those for Rio (2016) and London (2012) which in turn were double that of Beijing in 2008.

The 'cheapest' Olympic Games of recent times was Los Angeles in 1984 at 'only' $0.79b. LA's costs were comparatively cheap because, as the only bidder, they could rely on existing stadia and infrastructure rather than having to promise to build lavish new facilities to impress the International Olympic Committee decisions makers.

The rewards - more medals

The most obvious reward for a host nation is the uplift in medals as a result of increased investment in facilities and from a performance boost derived from athletes competing on home soil.

We've analysed the medal haul for the most recent hosts and compared it with their medal total in the games immediately before and immediately after the one they hosted. A hosting advantage would be reflected in an increase in medals, particularly gold ones, in their host year compared with the number they achieved in the games either side of their hosting.

The graphic shows the before, at home and after medal haul for the Olympics in South Korea in 1988 to Rio in 2016. Obviously, we can't include the Tokyo games in the after analysis but Japan got 17 more medals than in Rio and achieved 27 golds compared with 12 in 2016.

The chart shows that the host nation increased its medal haul from the previous games in all but one Olympics. The USA in 1996 got fewer medals than in 1992 although they did get more golds. In fact, every host nation got more golds compared with the number it achieved in the Olympics immediately before they hosted. One of the explanations could be that a nation sends more athletes to its home Olympics and would, therefore, have a better chance of winning more medals. In fact the data suggests that sending more athletes does not necessarily mean more medals. There are plenty of examples through history of nations achieving better with fewer athletes and worse with more.

The chart also shows that many host nations experienced a drop off in medals in the games immediately after hosting. The exception to this is Team GB that picked up more medals in Rio than they did in London (although they achieved slightly fewer golds) and Brazil that picked up two medals more than that achieved in 2016. Interestingly, Team GB's increased medal haul was achieved with fewer athletes than they took to London 2012.

The Olympic legacy

One of the central planks of the London 2012 bid was that the Olympics would inspire a generation to actively participate in sports and physical activities. Similar claims were made for the Olympics in Greece, Australia, China, Canada and Brazil. Did they deliver?

The answer is yes and no. Weed et al undertook a comprehensive literature review of the participatory outcomes of Olympic games and major sports events to assess whether there was, what they call, a demonstration effect, whereby people are inspired to participate in sport by such events.

They applied the outcomes in participation to a behaviour change model (BCM). We also use a BCM to assess where people are on a five stage pathway that goes from not being interested to being fully committed.

The research found that there was some evidence (although it was weak) to suggest that there is a demonstration effect particularly in the run up to the Games. This effect was must likely to be seen through existing participants either playing sport more often or switching to new sports.

It was also found that existing sports participants responded positively in behaviour to big sporting events hosted in other countries. Again this link was quite weak but it was there nonetheless.

There was no evidence in the research that major sports events encouraged new participants. There was, however, some evidence that previously lapsed participants came back to sport.

In relation to the BCM, the Olympic Games have more of an impact on those already engaged with sports (Planning, Action and Maintenance stages) rather than at the more attitudinal Pre-Contemplation and Contemplation stages.

Their conclusion was that big sporting events do not inherently inspire new participants to sport. Things are, however, a bit more complicated than this. Other research has shown that programmes that encourage people to be more active and grassroots initiatives for individual sports have resulted in increased participation off the back of an Olympics. Some of this is explained through people switching sports but it is also the case that a sport's exposure on a global stage combined with grassroots initiatives can start to change behaviours around new participation.

The implications for sports participation in the UK

Sports in the UK will want to build upon the exposure given by the Olympics. The big challenge is to address the more attitudinal barriers that stand in the way of behaviour change. The evidence shows that big sporting events on their own are not enough. Instead, sports need to adopt strategies that builds upon the global exposure by targeting growth at a local level.

This is where research is vital to the success of a growth strategy. We have set out a four point research plan to help sports develop effective growth strategies.

Assessin g attitudes and perceptions of the sport is essential. Sports need to understand how the public views it, especially how it is perceived in relation to others, how it fits into their lifestyle and the actions, events and messaging that will drive people to become aware, interested and engaged by the sport.
Sport will also need to understand drivers of participation. What are the push factors that will get more people involved and how can barriers to play be addressed? - for example, access, facilities and availability. Understanding what motivates people and how the sport can make an emotional connection is the key to sustaining participation.
The importance of schools in promoting interaction is key to long term growth in participation. Is your sport played in school and if not why not? What can be done to get more schools playing?
Inclusivity in sport is vital. Does your sport appeal across all community groups? What can be done to broaden its appeal?

The Tokyo Olympics has come after almost 18 months of lockdowns and restrictions. The spotlight is well and truly on sport. Now is the time to grab the opportunity.

If you'd like to know more about how our research can help your plans for growth, please get in touch.

Just another day

Thu, 29 Jul 2021 11:59:04 GMT

No dramatic change in the way we go out and about following 19th July 'Freedom Day'.

On 19th July the government scrapped all remaining restrictions in England that required masks to be worn in shops and other indoor settings, along with capacity limits in bars and restaurants and rules limiting the number of people who can socialise together.

On the same day, Scotland moved to Level 0 which eased social distancing rules, increased the number of people who could meet indoors and outdoors and allowed hospitality to open for longer. Employees were asked to carry on working from home where possible.

In Wales a more limited lifting of restrictions took place on 17th July and some restrictions on meeting indoors and outdoors were lifted in Northern Ireland on 26th July.

What was the upshot of the lifting of restrictions?

We used Google mobility data to map the impact. Google uses (opted in) location data from mobile phones to analyse the destination to which people have travelled.

The graphics show that Freedom Day was a bit of a damp squib. Comparing the Friday or Saturday prior to Freedom Day with the Friday and Saturday after 19th July shows very little changed, at least at the UK level.

Fewer people travelled to work on the Friday after Freedom day than before it. This might have something to do with the start of the school holidays in England. In Scotland, some regions saw an increase in travelling to work despite the request to carry on working from home if possible.

Fewer people went to the supermarket post Freedom day than before it. Again, some regions in Scotland and Northern Ireland bucked this trend.

Public transport was slightly down across the UK.

Freedom Day had more of an impact on travel to retail and recreation destinations (shopping centres, cafés, restaurants, theme parks, museums, libraries and cinemas). Travelling to these places was up almost 4 percentage points on the previous Saturday with South West and Central England, parts of Scotland and Northern Ireland recording above average travel. The figures for Northern Ireland are interesting because restrictions had not been formally lifted at this point.

Travel to destinations other than the supermarket or pharmacy are still well down on pre-Covid levels. This probably reflects people's inherent caution as infection rates were rising at the time.

The data shows that despite Freedom Day we are still a long way from returning to normal.

We need to talk about vaccine passports

183:752711313 (Michael Wagstaff) — Tue, 20 Jul 2021 17:05:09 GMT

We identify six personas in the debate. Which one are you?

Recent research by YouGov suggests that the majority of Britons would support the introduction of a 'vaccine passport' that people who have had the vaccine could show to gain entry into places and events.

Overall, 58% support the idea of a vaccine passport in principle. Looking at specific uses, 56% supported its use for entry to bars and pubs, 48% for visiting the hairdresser and 44% for entry to a coffee shop.

The government has signalled its intention to make vaccine passports mandatory for entry to nightclubs and other crowded spaces at the end of September. In the meantime, it has said that private businesses can ask customers about their Covid status if they wish as long as they do not break the law.

Six degrees of separation

While surveys show a broad level of support, they mask a strong debate about the pros and cons of vaccine passports. By analysing thousands of tweets on the issue we have tapped into the national conversation to shine a light (and at times a little light relief) on what Britain really thinks about vaccine passports.

Our analysis shows that there are six different personas in the debate, each having very different opinions. Some of these are rooted in deeply held beliefs. Others are just odd.

Which one are you?

The Freedom Fighter

The most vocal of the bunch, the Freedom Fighter puts personal freedoms front and centre of the debate. But this is no 'woke' lefty libertarian defending our basic human rights. No, the Freedom Fighter is concerned more about defending the economy by upholding British values of never surrendering, being free and fighting the enemy. War imagery peppers the debate but so too does the flag waving patriotism and nationalism that has become more prominent recently in politics.

Freedom Fighters are not just against vaccine passports; they are against all Covid restrictions. The risks to the economy and society (but mainly the economy) are too great to carry on as we are. Now is the time to open up and get back to normal. Dunkirk spirit and all that.

The Ethical Objector

The Ethical Objector might be on the same side as the Freedom Fighter but they are a very different person. The Ethical Objectors argue that a vaccine passport is discriminatory and divisive. It pitches those who haven't got a vaccine against those who have. It pitches young against old. It divides the vaccine hesitant and those who distrust medical decision makers from the rest of society.

They argue that vaccine passports discriminate against those who cannot have the vaccine due perhaps to illness, pregnancy or being unable to leave home. How is it fair, the argument goes, that these members of society are excluded?

There is a further line of argument that states if staff in health and care are required to have had a jab then this could discriminate indirectly against women and staff from an ethnic minority backgrounds who tend to form a significant proportion of workers in this industry.

Specific and Limited

Members of the Specific and Limited group hold a view somewhere between Ethical Objectors and Freedom Fighters. They do not believe that people should have a vaccine passport to go to the pub or hairdresser or to do anything that is part of everyday life. They believe, however, that vaccine passports will be necessary but limited to specific things: international travel and frontline health and care staff.

They take great issue with Ethical Objectors when it comes to frontline health and care workers. Health workers, they argue, have a duty of care both to their patients and themselves. This group use reports of some vaccine hesitancy among health workers as evidence of the need for passports to prove that staff have been vaccinated. The passport is proof that they are protecting the most vulnerable from the virus as well as themselves.

There Is No Alternative

TINAs on the one hand have sympathy with the arguments of Ethical Objectors but on the other hand reject them. Yes, they say, there are ethical considerations and a risk of division but vaccine passports are a necessary thing. If we want to get back to some form of normality then the passport has to be introduced.

In backing up their arguments TINAs point to Israel's 'green pass' system. The 'green pass' is a smartphone app that records whether the user has been vaccinated, has had a recent Covid-19 diagnosis meaning that they will have antibodies in their system or a recent negative test. If any of these conditions are met they are allowed into venues including gyms, swimming pools, restaurants and cafes, hotels, sports venues, theatres, cinemas and exhibitions.

TINAs want the 'green pass' introduced into the UK as soon as possible to get life back to normal. After all, there is no alternative.

Conspiracy Theorists

Almost matching Freedom Fighters for Twitter noise, Conspiracy Theorists are a mistrusting lot and probably spend too much time on the socials.

Conspiracy Theorists are superspreaders of views that are a little bit outside the orbit of normal people. For example, Covid-19 is a bioweapon deliberately let loose by the Chinese government. Alternatively, Big Pharma RELEASED THE VIRUS TO MAKE BILLIONS FROM VACCINES. Oh yes, they use capital letters a lot. Oh and of course vaccines don't work and in fact cause more harm than good because some bloke who is at Harvard, or went there, or at the very least lives nearby IS AN EXPERT AND HAS PROOF.

Governments only want to introduce vaccine passports to control and monitor their citizens. Covid doesn't exist and if it does is no worse than flu. You don't need to wear a mask. And so on and so on. And breathe

Dedicated Followers Of Science

Dedicated Followers Of Science know all the facts. They know their R from their elbow. They know the difference between an mRNA vaccine and an inactivated one. They have become truly expert in the detail of infection rates, vaccination rates, lockdown roadmaps and Chris Whitty.

But what do they think of vaccine passports? Next slide please.

Home or Rome?

Sun, 11 Jul 2021 14:23:42 GMT

Who needs Ronaldo and Mbappé when you have the big lad Data Science in your team?

It's the Euro 2020 final and we have cranked up our predictive model one more time. Our model analyses form coming into the Euros and results at the tournament to calculate each team's probability of winning.

Is football coming home or is it going to Rome?

According to our predictive model it's incredibly close and as near a 50-50 game as can be. England's home advantage just shades Italy's superior form to give them a wafer-thin edge.

This is the closest of all the match probabilities that we have modelled throughout Euro 2020. Expect a tense and nervy affair.

The finalists' story

Looking back at the model we can see that both England and Italy were overwhelmingly favourites at all stages to qualify from their groups.

Both teams benefited from the draw but England more so. Italy were drawn on the harder side but had easier opponents in Austria in the round of 16. Italy had a 75% probability of winning that game although they needed extra time to do so. England were drawn on the easier side playing their potentially most difficult fixture, Germany, at home. England had a 53% probability of beating Germany which was their lowest probability of winning any match up until the final.

Given that the model predicts a more or less 50-50 game it's worth looking back in history to see who has the upper hand in meetings. England has never beaten Italy at a major tournament. In the 2014 World Cup, Italy won 2-1. At Euro 2012 the score was 0-0 with Italy winning on penalties. At Euro 1980, Italy won 1-0. History favours the Azzurri.

What about the dreaded penalty shootout? Italy has faced penalties 10 times in major competitions (World Cup and Euros) winning four times. All four of their wins have come in the last six competitions.

England has faced penalties eight times in major competitions, winning twice. England fans will be hoping it doesn't go to penalties.

How has data science done at the Euros?

As we reach the final match in Euro 2020, it's time to review the performance of our predictive modelling. Did it win the golden boot for accuracy or did it let the team down and let itself down?

Before a ball had been kicked we calculated that finishing the group stage 3rd with 3 points and a goal difference of no worse than -1 would be sufficient to qualify. We gave it a 67% probability and this is exactly the combination that Ukraine achieved to go through as the 4th ranked 3rd place team.

After the 1st round of group matches we calculated that there was a 71% probability of Italy, Belgium, Netherlands, England, Spain, France, Portugal and Germany all qualifying for the knockouts. All did qualify despite both Germany and Spain failing to win their opening matches.

Ahead of the final group games we revisited the probability of finishing in 3rd place and qualifying. The model calculated that the four 3rd place qualifiers would come from groups C, D, E and F . In fact they came from groups A, C, D and F. The model calculated that there was a 56% probability of the 3rd place qualifier coming from group A and 60% probability from group E. Although the model was wrong it was still very close.

There were some shock results that went against the probabilities in the round of 16 . Czech Republic was given a 29% chance against the Netherlands and Switzerland a 34% chance against France.

In the quarter finals, the model calculated that Belgium would beat Italy but the probability was 52%-48% so Italy winning was no real surprise. All of the remaining quarter final and semi final matches went with the model's probability scores.

Overall, the model done great and more than justified its selection.

You can catch up with all our Euro 2020 probabilities from throughout the tournament here .

Scrapheap Challenge

Sun, 04 Jul 2021 16:33:16 GMT

The over 50s have been hit hard by redundancy during the pandemic.

Three people tell us their story and the challenges they face.

What a waste

Fri, 21 May 2021 12:03:28 GMT

Households in the UK throw away £19 billion worth of food each year

The electric bike revolution

183:752711313 (Michael Wagstaff) — Fri, 07 May 2021 16:47:16 GMT

Sales of e-bikes set to go up a gear

There is a lot of discussion about how electric cars will change the automotive landscape but not so much about the impact of electric bikes. Imports of e-bikes in the last few years, particularly from China, have increased rapidly and so growth in use seems inevitable.

We've brought together some of the more interesting stats about e-bikes in this handy infographic. If the forecasts are right then e-bikes will become a much more familiar sight on the road.

Research shows that e-bike owners cycle longer distances than riders of normal bikes. This means that there is a greater potential for e-bikes to replace some journeys normally undertaken by car. In fact, one quarter of owners say they use their e-bike for journeys instead of the car so the environmental impacts are already becoming evident.

Despite the bold predictions of growth, only 7% of people in the UK say they want to try an e-bike in the next 12 months. In Italy the corresponding figure is 30%. It is clear that significant investment in UK cycling infrastructure will be needed to achieve the level of growth forecast.

If you don't want to know the score look away now

Fri, 30 Apr 2021 15:30:03 GMT

Using predictive modelling to forecast the Premier League final table

With the Premier League season down to the last few games, predicting the final table should be easy, right?

While it is true that Manchester City look nailed on for the title and the current bottom three (Sheffield United, West Brom & Fulham) look all but down, there is actually still much to play for. In fact, even at this late stage, there are still many possible permutations depending on how results go.

Many of us have done that thing where you look at your clubs remaining fixtures to predict which ones are winnable, which ones you'll struggle to get anything from and then see where your predictions leave you in the table. We like doing that too but with a bit of a data science twist.

We've had a go at predicting the final Premier League table using data modelling. In 1997, Dixon and Coles presented a model for predicting football scores by modelling the number of goals scored in a match as a Poisson distribution (the probability for a team to score one goal, two goals, three goals and so on in a match). We've applied this model to the remaining Premier League fixtures.

To generate the prediction we fed in the results so far from this season's Premier League. The model sets parameters to represent a team’s attacking strength, defensive strength and home advantage. The model also contains a time decay component so that more recent results are given a higher weighting. Why is this? You may remember a few seasons ago that Crystal Palace lost their first seven games of the season. A change of manager saw a sharp upturn in Palace's form and they finished the season well clear of the drop. Had the model not down weighted the earlier poor form, it would predict worse results for Palace than that which they actually got.

The model generates a range of probabilities for each possible result (based upon each teams probability of scoring one goal, two goals and so on) and assigns 3 points for a win and 1 for a draw to their current points total.

You can see the range of probabilities generated by the model by looking at the possible scores matrix for the Southampton v Leicester game.

We ran the model multiple times and then took the average of all iterations to produce the final table you see below. The model calculates total points for each team and generates probabilities of finishing in certain places.

Our table shows teams ranked by predicted points along with the probability of finishing in that position. It also shows the probability of the team finishing higher than the position predicted or indeed lower.

So what does it show?

Campeones, campeones, olé, olé, 0lé

No surprises here. It's Manchester City.

The model suggests that City are 99.99% recurring likely to be top of the league.

These are the champions (League places)

Manchester United has a 90% probability of finishing second and 10% probability of finishing third or fourth. They look certain to join Manchester City in the Champions League next season.

Leicester is predicted to finish third with a 69% probability of doing so. It is possible that they could get second (10% probability). On the other hand, there is a 5% probability that they will finish outside the Champions League places.

The race for the final Champions League place is less clear. The model suggests that Chelsea will snatch the 4th place trophy. Fans of West Ham, Liverpool and Spurs shouldn't give up their Champions League hopes just yet - there is a 53% chance that Chelsea will finish outside the top 4.

Of course Arsenal, who we predict to finish 10th, could make the Champions League by winning this year's Europa League.

We're gonna win the (Europa) league

The fifth place team goes directly to the Europa league group stages. A place is also given to the FA Cup winners with the winner of the EFL Cup going into the newly formed Europa Conference. If, however, the FA Cup or EFL cups are won by a team that qualifies for European football due to their league position, then the 6th place team gets a Europa League place and the 7th placed goes into the Europa Conference. As the FA Cup final is between Chelsea and Leicester and Manchester City have already won the EFL cup then this is likely to happen.

The model predicts that West Ham and Liverpool will qualify for the Europa League and Spurs (in 7th place) will qualify for the Europa Conference. But don't give up just yet Everton fans. The model predicts that there is still a 40% chance that the Toffees will qualify for Europe. Game on as they say.

Going down, going down, going down

Fulham and West Brom are set to join already relegated Sheffield United in the Championship next season. Although the model predicts Fulham will be relegated, there is still a 5% probability of them avoiding the drop. It's the hope that gets you.

Big Sam (Allardyce) has never been relegated but the model predicts his West Brom team will finish second bottom.

VAR checking for possible unknowns in the prediction

Our Premier League table is based on a predictive model. With any predictive model there will be things happening that are not taken into account. These include the impact of player availability on form. A side might lose form with the loss of a key player or gain form when a key player comes back into the side.

Sometimes teams with nothing left to play for play with a new found freedom that turns the form book on its head. Similarly, teams whose season is effectively over might be, metaphorically speaking, on the beach. Others, seeing the chance of qualifying for Europe or avoiding relegation play much better than they have recently.

And of course VAR could rule that an armpit hair was in an offside position thus denying a vital goal that could have a profound effect on the table.

Predicted Premier League final placings 2020/21

For some teams there is a good chance that they could finish higher or lower than their predicted position. The graphic below illustrates just how much things can change for individual teams.

In the graphic we show the range of possible placings. The darker the shading the stronger the probability of finishing in that place. For example, Burnley could finish anywhere between 11th and 17th but 14th is the most likely. Leeds could finish as high as 8th and there is a set of results that see Spurs finish 4th.

For many teams there is still all to play for.

England goes shopping and drinking

Tue, 27 Apr 2021 14:45:35 GMT

Having fun makes welcome return

We have analysed the latest Google Community Mobility data to see the impact of the lifting of restrictions on non-essential shopping, pubs. cafes, restaurants and leisure activities in England. Google's data is fascinating. They use (opted in) location data from mobile phones to plot where people have gone.

We compared where people went the weekend before the lifting of restrictions in England on 12th April with where they went the weekend after. We see huge increases in the number of people going out and about, doing some shopping, having a drink and a meal and having fun.

Lots of people went to Lakeside shopping centre in Thurrock. Not necessarily everyone's idea of fun.

Things can only get bitter

183:752711313 (Michael Wagstaff) — Fri, 05 Mar 2021 18:54:59 GMT

The public's focus on Covid-19 end game rather than blame game is bitter pill for Starmer to swallow

I can remember it as if it were yesterday. John Prescott and Peter Mandelson jigging awkwardly to D:Ream's Things Can Only Get Better. It was the day after the 1997 General Election. A Labour government had been returned after 18 years in opposition. Optimism was in the air. Tony Blair's Labour Party seemed unassailable.

Fast forward 24 years and Sir Keir Starmer's Labour is in a very different place.

A recent YouGov poll gave the Conservative's a 13 point lead over Labour. The Tories lead despite accusations of dithering over lockdown, PPE shortages at the start of the pandemic, allegations of cronyism over government contracts, the government held by the courts to have broken the law, problems with Test and Trace and over 120,000 dead from Coronavirus.

Labour party supporters must be scratching their heads wondering what on earth is going on. How can a government that has performed so poorly during the pandemic be so far ahead in the polls?

I think there are two things going on. First, the British government is not alone in being criticised for its performance. YouGov tracking data shows that significantly less than half (43%) of Britons think their government has performed well during Coronavirus (as at 22nd February 2021). They are not alone. The French government is tracking below the UK in relation to approval by its citizens. It's a similar story in Spain. The USA has only moved slightly ahead of the UK on public approval since the election of President Biden. Approval for the German government is higher at 48% but has tumbled sharply in recent weeks as the vaccine roll out has stuttered.

People see that governments of some of our closest allies are struggling too and so give the Tories the benefit of the doubt. Further, data from Opinium suggests that voters are not convinced that Starmer would have handled the pandemic any better. Again, this explains why Tory support in the polls is so high.

The other issue is that with the success of the vaccine roll out people are now focused on the end game - getting back to normal - rather than the blame game. The vaccine has brought the end into clear view hence the bounce we see in the polls. Analysis shows that there is a relationship between when people think life will return to normal and Conservative support.

We've used YouGov's voting intention tracker and plotted it against ONS data on the percentage of people who think that restrictions will be over sooner rather than later (measured by the percentage saying normality will return in six months or less).

Figure 1 : Percentage intending to vote Conservative plotted with percentage who think we will be back to normal in 6 months or less

The graphic shows the effect of the link over two time periods. The period from April 2020 to October 2020 saw a downward trend in the percentage of people who said normality would return sooner rather than later. We can also see a downward trend in Conservative support. Conversely, from October 2020 to now we have seen an upward turn in the percentage who think normality will return within six months. We also see an upward turn in the Conservative's popularity.

In fact, voting intention and thinking that things will return to normal in less than 6 months are reasonably well correlated (correlation coefficient of 0.6, where 0 is no correlation and 1.0 is a perfect correlation) and helps to explain the polls.

It seems that none of the negatives over the government's performance in the last 12 months have seriously dented Tory support. Now that the vaccine is being rolled out and normality is tantalisingly close, the Conservatives are getting a bounce.

For Sir Keir Starmer, the public's focus on the end game rather than the blame game must be a bitter pill to swallow.

12 months of Covid-19

Fri, 19 Feb 2021 09:49:52 GMT

12 months of Covid-19 has had a huge impact on how much we go out and about

Growing the charity brand

183:752711313 (Michael Wagstaff) — Wed, 17 Feb 2021 14:48:05 GMT

Using brand assets to stand out

These are tough times for charities. Public donations are down and many fundraising events have been cancelled. The sector is predicted to have lost £12bn in income in 2020.

Standing out from other charities by building mental availability and ensuring maximum physical availability online and offline has never been more important. It is the key to growth for charities large and small.

In this publication we discuss how the look, tone and feel of brand assets, such as logos, colours, taglines and images, can drive mental availability. We also discuss having an online presence, acquiring and retaining supporters and the role of marcomms in helping charities grow.

You can view the publication online below or you can download it by clicking here .

Covid 19 Social and Economic Indicators

Mon, 01 Feb 2021 13:04:22 GMT

Data as at 20th January 2021

Football's gone home

183:752711313 (Michael Wagstaff) — Sat, 16 Jan 2021 11:42:51 GMT

Our research proves that the crowd really does make a difference to the home team's performance.

And the winner is!

Mon, 21 Dec 2020 20:08:20 GMT

How we did using sentiment analysis to predict Strictly outcomes

Strictly Come Dancing ended on Saturday with the grand final. We used sentiment analysis to predict that Bill Bailey would win the Glitterball, which he did. Champagne all round.

We also tweeted an updated sentiment score after each round of dancing which showed Bill stretching his lead as the show progressed.

But it wasn't just for the final that we used sentiment analysis on Strictly. Over the last eight weeks we have analysed tweets to predict the outcome of each show. So how did we do?

How accurate is sentiment analysis?

Overall, we think our Strictly Sentiment model performed really well. We got 13 out of 15 measurable outcomes correct - an 87% success rate has got to be very good in anyone's book.

Sentiment analysis is the process of determining whether some text is positive, negative or neutral in tone. To extract sentiment from tweets we built a supervised machine learning based sentiment model. To train the model we manually labelled thousands of Strictly tweets from the launch show as negative, neutral or positive and fed these into it. We paid particular attention to teaching the machine how to deal with nuance and ambiguity. For example, the sentiment behind the word 'poor' in this sentence: 'Poor Jacqui having to dance with Anton', is very different to: 'That dance was poor Jacqui'.

Our work was successful as the 87% success rate demonstrates.

Each week we grabbed thousands of tweets about each contestant. When then ranked them on sentiment and gave points depending on their rank. This is the same method that the BBC uses to give each contestant a score based on the public vote. We then combined the ranked sentiment score with the judges score to predict which two contestants will be in the dance off. The prediction was tweeted one hour before the results show on Sunday. The exception to this was for the semi-final when we tweeted the prediction a few minutes after the show. For the semi-final we moved to analysing the tweets in real time in preparation for the final.

Out of 14 dance off places we got 12 right. The two times we got it wrong was when we put JJ Chalmers in the dance off. We put this down to JJ having a significant offline following. This was the one weakness of the model.

For the final, our analysis predicted that Bill would be the winner. We tweeted this prediction just after the public vote closed and 10 minutes before the live announcement on tv. Obviously, we were really pleased when Bill was announced the winner but doubly pleased when we saw that two online viewer polls had HRVY ahead of Bill.

O ur sentiment model had outperformed an established online polling method. Fab-u-lous!

Highs and lows

There have been a lot of high points for us:

Correctly predicting 13 out of 15 outcomes.
Building a sentiment model that is highly accurate and that outperformed established online polls.
The model was sensitive to movements in popularity - Maisie going low then high, Ranvir going in reverse.
Undertaking analysis in real time was a definite squeaky bum time win.
Seeing traffic to our website sky-rocket at the weekend and when the weekly post analysing the results was published.

But it's not all been good:

The model was less effective for contestants whose following was more offline than online.
People tweeting about the show were younger than those who watch it (although the audience data on which we made this comparison in a previous post is a few years out of date).

Can sentiment and text analysis be used more in market research?

Yes it can.

The most obvious example is analysing customer feedback, either using a brand's own feedback channels, open-ended survey responses or utilising social media. Sentiment analysis can be used to keep track of how a brand is performing with the public or potential buyers. Topic analysis (making sense of unstructered text) can be used to understand what consumers are talking about. Gaining valuable Voice of the Customer insights from the application of sentiment and text analysis is a clear business advantage for brands.

We use sentiment analysis in combination with topic analysis to identify themes and subjects from large unstructured data sets. We have posted quite a few examples of topic analysis in the Scribbles section of our website where we gauge public opinion on events and issues. For example, we produced an analysis of the nation's mental health during lockdown restrictions by analysing 15,000 tweets.

We also undertake sentiment and topic analysis to analyse ratings and review sites. We do this at brand and product level. We scrape sites such as Trust Pilot and Amazon to identify strengths and weaknesses of product offerings and those for rival brands. This analysis can also feed directly into new product development by identifying gaps in product features (compared with rivals) and opportunities in the market based on unmet customer needs.

Strictly Sentiment leaderboard

Our final Strictly Sentiment leaderboard showed a win for Bill. He really does love to boogie on a Saturday night.

And so do we.

S trictly Sentiment score

Visits to shopping centres down by a quarter compared with Feb 2020

Mon, 14 Dec 2020 12:32:14 GMT

Travelling for retail or recreation at 74% of its Feb 2020 baseline.

We find Google's Community Mobility reports fascinating. They use (opted in) location data from mobile phones to plot where people have gone. The reports look at visits to retail, transport hubs, supermarkets and places of work and compares these with a median value from Jan 3 to Feb 6 2020 (the Feb baseline) to calculate the percentage drop in visits.

The graphic shows the drop in visits to retail and recreation (shopping centres, cafes, museums and the like) from the baseline to 7th December 2020.

The darker the shading the nearer the number of visits is to the Feb baseline. The lighter the shading the bigger the drop off.

Three of the biggests drop offs are in Scotland. Stirling and Glasgow areas have seen 50% drop off in what should be a very busy time. London is down 38% from the baseline. Greater Manchester 28%. Kent (also in tier 3) is only down 26%.

If you want normality go to the Isle of Wight. It's currently running at 90% of its Feb baseline.

The Twelfth Man is not a myth. Fans really do improve the performance of their team

Thu, 03 Dec 2020 11:04:16 GMT

Home performance affected by lack of football crowds

If you don't want to know the score look away now

Fans have missed football and football has missed fans. The crowd is often referred to as the '12th man', helping the team home and roaring them to success. The crowd is thought to lift performance by getting behind the players. How can a team not be inspired when thousands of fans are singing and chanting your name?

The sound of the crowd has been missing since the first lockdown in March. When football restarted in the summer, no fans were allowed into the ground. Matches have been played against a background of coaches shouting from the touchline and the odd swear word on the pitch. Some clubs, such as Gillingham in League 1, have pumped fake crowd noise into the ground to improve the atmosphere and help the players.

The big question is does the lack of a crowd affect performance and are some teams more likely to be affected than others?

To find out we analysed results for matches played without a crowd from June 2020 to end of November 2020 and compared them with matches with crowds from previous seasons. We found that:

In the Premier League, the percentage of home wins for all 190 matches played without a crowd is low but not the lowest in recent history.
The percentage of home losses, however, is historically high at 38%.
As a result, there is a significant difference in the percentage of points won at home without a crowd than with. The proportion of points gained at home without fans is 52% compared with 57% for the last completed season with fans. For the 2020/21 season so far, only 46% of points have been gained at home by Premier League teams.
The impact of the lack of crowds diminishes the further down the football league we go. Performances in Leagues 1 and 2 are the least affected by the lack of crowd as average attendances were lower to start with.

The pre-match build up

The Premier League and the EFL were both halted in March 2020. The Premier League restarted in June and completed 92 matches. The Championship also restarted in the summer but Leagues 1 and 2 did not complete the season with final placings being decided on a points per game system.

Much of our analysis is focused on the Premier League. The average attendance in the Premier League is much higher than in the EFL which means that the impact of having no crowds is likely to be greatest.

To analyse the impact of having no crowd we have combined the 92 matches from the end of last season with the 98 matches played this season to 30 ^th November 2020. In total there are 190 matches across two seasons where no crowd has been allowed.

Some analysis is included of the EFL but this is based on this season’s results and therefore contains fewer matches. It is, however, a useful indication of the impact on the EFL.

Kick off

We kick off by comparing the average number of wins by the home side in the Premier League with and without crowds. To do this we have used the average percentage of home wins in each season from 1993/4 to 2018/19 compared with the 190 Premier League matches for which no crowd was allowed (denoted by the red X ).

Figure 1 shows that although the proportion of home wins is historically low without crowds it is not abnormally low. For the lockdown games 42% ended in a home win. In the last full season before lockdown the percentage was 48%. A lower percentage of home points was recorded in the 2015/16 season (41%).

Figure 1 : Percentage of home matches won, drawn and lost and points won at home

It’s the same finding when we look at draws. The figure for the lockdown games is historically low but not abnormally so.

Where we see a big difference is with the percentage of losses. The 38% home losses for the crowdless games is historically high and this translates to an historically low percentage of points gained from home games. In the last full season before lockdown, teams, on average, achieved 57% of their points total at home. In the 190 games without fans this figure fell to 52%. This means that there is virtually no home advantage when no crowd is present.

At this point we have to ask how much of the drop off in home advantage is due to the form and relative strengths of the teams that have played at home this season? In other words, is this just a quirk of the fixture list pitching an out of form or weaker home team against stronger more in form opposition? The answer to this is no.

In 1997, Dixon and Coles presented a model for predicting football scores and results by modelling the number of goals scored in a match as a Poisson distribution (the probability for a team to score one goal, two goals, three goals and so on in a match).

Their model identified parameters to represent a team’s attacking strength, defensive strength and home advantage. Applying this model to Premier League seasons from 1993 and weighting more recent results to represent current form, we can obtain a measure of home advantage independent of the current attacking and defensive qualities of teams.

Figure 2 : The home advantage parameter

The second half

How does the lack of crowd affect the EFL compared with the Premier League? Using only data from the 2020/21 season up until 30th November 2020, we can see that the percentage of points won at home by Premier League teams falls to 46%. This is 12 percentage points down on the average points gained at home in the last five years.

The percentage of home games won in the Championship so far this season is three percentage points down on the average for the last five seasons. For League 1 and League 2 the drop off in points picked up at home is less than two percentage points down on the last five years.

Given that the lack of a home crowd diminishes performance, it is not surprising that there is negative correlation between crowd size and the percentage of points won at home, as shown in Figure 3. The higher the average crowd, the greater the drop off in home performance.

Figure 3 Relationship between average attendance and percentage drop off in points gained at home

Post-match analysis

Crowds play an important part in football and the lack of them has led to a diminishing of home advantage. With no ‘twelfth man’ to cheer on the team, the home side has lost its edge. Does this drop off in performance affect some teams more than others?

Figure 4 suggests that it does affect some teams differently. For this analysis we have calculated a home points per game (PPG) differential score. This is the number of points per game each team gained at home without the crowd minus the PPG with crowds.

For example, if a team gains 1.5 points per game at home with a crowd and 1 without a crowd, the difference is calculated as -0.5. The PPG with crowds is derived from match results from August to March of the 2019/20 season. The no crowd PPG is derived from results from the rest of the season plus the 2020/21 season to November 30 ^th 2020. The chart excludes results for the three teams relegated at the end of last season and the three teams promoted.

Figure 4: Points per game differential (PPG without crowds minus PPG with crowds)

The chart shows that over half of Premier League sides gain fewer points per game at home without a crowd. It should be acknowledged that for some sides a loss of form might account for them gaining fewer points. Sheffield United for example have been poor at home and away this season. Liverpool is another. Although Liverpool is still very good they are not at the same heights as previously. Even so, it shows that some clubs have been hit worse than others by the exclusion of fans.

Our analysis shows that crowds really do make a difference to their team. This is most pronounced in the Premier League where crowd sizes are much higher. The return of home fans will be of considerable benefit to some teams.

Teams whose fans are not allowed back into the stadium might be at a disadvantage compared with those whose fans are allowed back in. Our research suggests that there might not be a level playing field for all teams as a result.

Are house prices making the right move?

183:752711313 (Michael Wagstaff) — Tue, 20 Oct 2020 13:58:43 GMT

Ever increasing house prices are good for some but not for others

According to the property website Rightmove, the price of properties coming onto the market is at a record high. An increase of £3,500 in September saw the average list price grow to £323,500. Year on year prices have increased by 5.5% which is the highest annual growth for four years.

Rightmove is now forecasting that the average asking price will increase by 7% in the year to December. They called their previous forecast of a 2% increase 'timid'.

Higher asking prices are being fuelled by the release of pent-up demand following lockdown, the stamp duty holiday on properties up to £500,000 and people making a lifestyle choice to move away from cities for more space in rural areas and the coast.

While rising house prices are great news for some people. But for others they are very bad news as they push up the cost of home ownership and take it out of reach altogether. As a result many young people, couples and families find their housing choices limited, their economic mobility reduced and mortgage repayments unaffordable.

Housing affordability is a huge problem in certain parts of the UK.

My rule of thumb is that anything up to 25% of net income (take home pay) is an affordable housing cost. On this measure large parts of the UK are unaffordable for first time buyers.

And for some regions the numbers are scarily unaffordable as the chart shows.

FTBs in London are handing over half their disposable income to pay the mortgage. Buyers in the south east and south west are spending over one third of their income.

Some might argue that there is a lot to be said for a housing market where prices are 'timid' especially if it encourages greater choice, greater affordability and greater quality.

High house prices and low affordability are just not sustainable in the long term. Isn't it time for a rethink on what society wants from housing before prices move right out of reach?

We need to talk about students

183:752711313 (Michael Wagstaff) — Wed, 07 Oct 2020 12:00:27 GMT

Locked up and let down or paying the price for partying? Britain talks about students.

Students have returned to Britain's universities and predictably this has been met with a spike in coronavirus infections.

Outbreaks have been seen at many universities but most notably in Glasgow, Edinburgh, Manchester, Liverpool, Newcastle, Northumbria and Swansea.

For many, particularly first year students, what should have been the start of a life-enhancing experience has instead turned into a nightmare of restrictions, lockdowns and having to isolate with people they barely know.

News media have carried stories of the outbreak, highlighting the restrictions and quite often the humour that goes with it - most papers carried photographs of banners asking for beer and other student essentials to be sent. Not everyone has been supportive though with quite a few vox pops blaming the students themselves for partying and ignoring restrictions on meeting up indoors.

What then is the sentiment towards students and the restrictions they face? To find out we undertook a topic and sentiment analysis of over 6,000 tweets (and retweets) that specifically mentioned universities and coronavirus restrictions. The tweets covered the period from 21st to 28th September when reporting of the university outbreaks was at its highest. We stripped out tweets that were not about UK universities but left in those that compared the UK situation with other universities around the world.

The analysis was undertaken in-house using our own text analysis tool. This uses natural language programming and machine learning to identify topics from large datasets of unstructured text. We then undertook a qualitative analysis of the data to get under the skin of what people are really talking about.

What are you talking about?

We identified four main topics of conversation:

The impact on the health and wellbeing including financial wellbeing of students.
The science behind the restrictions
Government Covid-19 policy and how it is communicated
International comparisons with other universities

The most talked about by far is the impact of restrictions on students' health and wellbeing as the graphic below shows.

Health and wellbeing

Over two-thirds of the conversation concerns the effect of the restrictions on mental health, anxiety and stress levels. Particular concern is expressed for first years who may have to self-isolate with people they have only just met. Commentators picked up on the uncertainty of what university life will be like this academic year. Learning may be more fragmented. Students might not be allowed home for Christmas, social interactions will be limited, the opportunity for friendships, shared experiences and being part of society may be curtailed. It was pointed out that many students work part-time in order to make ends meet while studying but jobs have gone in the pandemic, leading to increased financial worries.

It is felt that the current situation will have a significant impact on the mental health of many young people.

Another strong theme within this category is that of fairness both in relation to financial fairness but also a sense of a social injustice. Students (in England) are having to pay over £9,000 in tuition fees and take out student loans to cover accommodation and living expenses. Going to university is an expensive business and students should expect to receive value for what they are paying for. Allied to this is an argument that universities have been eager to see students return for the tuition and accommodation revenue they bring in.

It is thoug ht that being stuck in halls with lectures almost wholly online does not represent value and therefore many tweets call for students to be given refunds on their hall costs, tuition fees or both. It is pointed out that Glasgow is refunding some rent costs and other universities should take their lead.

There is also a sense that the situation with students represents a social injustice. When they signed up for university they signed up for the whole experience - learning, developing as a person, meeting new people, experiencing new things and yes socialising. The restrictions are having a negative impact and following so closely on the A level crisis in the summer, means some students are being hit hard twice.

Many tweeters talk about their own experience at university how the current student cohort suffers by comparison. Tweeters also point out that young people face a difficult job market and this, added to the debt from doing a degree, plus Covid-19 restrictions mean that it will not be a good time for young people into the foreseeable future.

It's not all supportive though. There is a view that rule breaking and lack of social distancing has led to restrictions being imposed. Students, the argument goes, only have themselves to blame. Similarly, some tweeters question why students have returned to universities in the first place. They must have known that cases would spike as a result. The sensible thing would have been to stay at home.

There is also a lot of humour from students and other commentators throughout the conversation. There is talk of remaking Die Hard where Bruce Willis has to rescue his daughter from university lockdown. References to emergency supplies of beer and food are made as well as middle class parents parachuting in Waitrose survival hampers.

International comparisons

The conversation contains some interesting comparisons on how universities around the world have reacted to the pandemic. There are some similarities to the UK in relation to the move to mainly online learning and the associated need for training and support for teaching staff to deliver education in a new way. Related to this, are examples where universities encouraged students, especially international ones, to stay at home and learn remotely. Learning has been augmented through the use of video tutorials, email contact and WhatsApp groups.

There were examples where universities were still doing some face to face learning but these had been scaled right back and strict social distancing measures introduced.

Other examples included having virtual campus tours rather than open days, allowing deferments for a year and channelling extra resources into outreach work to try to prevent students dropping out due to distress.

Lockdown science

Tweets in this topic quoted an interview with Professor Mark Woolhouse from Edinburgh University, who is also a member of Sage, the government advisory body. Professor Woolhouse said that lockdown was an unproven policy that only defers the spread of the virus rather than eradicates it. Reference was also made to academics from Harvard and Stanford universities who argue for age-related measures rather than total lockdowns.

Sweden came up a lot in the conversation and how their approach to the pandemic was different. Questions were asked about why the UK does not follow it, although it is clear that some people do not realise that the Swedish approach is more nuanced than their understanding of it.

Government policy

Tweets in this topic focus on the government's response to the pandemic in general and the university situation in particular.

There is criticism over perceived failures of Test and Trace, references to Dominic Cummings and his trip to Barnard Castle giving students a ready-made excuse to ignore restrictions, as well as more general references to 'chaos' in the government's response to the university situation. This relates to a perceived lack of planning and lack of clarity in communications. A clear sentiment in this category is that students have been 'hung out to dry' by the government (and universities) and that much greater financial and emotional support is needed.

The notion that students might be locked down for Christmas attracts criticism over its fairness and whether students and their parents will actually comply with it. Much of the blame for the 'crisis' is pinned firmly on Gavin Williamson (Education Secretary) and Matt Hancock (Health Secretary).

Key take-outs from the analysis

The analysis of tweets has shown the breadth and depth of the conversation on student restrictions at university. Much of what was found could have been predicted - the debate about mental and financial wellbeing, the effectiveness of lockdown and the criticism of the government response, are not a surprise. No surprise too is the sentiment that student actions contributed to the restrictions.

Not all findings are predictable though. The analysis shows how much people are prepared to engage in debate and highlight examples of good practice and experience from universities around the world.

One of the most unexpected findings is the sense from many tweeters (and not just students) that young people are on the receiving end of a social injustice. An experience that should be about self-discovery, expanding horizons, learning and independence has turned into a dystopian nightmare. The public seem to have latched on to this unfairness. Sometimes it is used as a stick with which to beat the government but for the most part people have shown genuine empathy and understanding towards young people.

In a society that is increasingly being defined along binary lines (left v right, old v young, us v them), it is good to see an example of people coming together to speak out against unfairness.

How Britain has changed

183:752711313 (Michael Wagstaff) — Thu, 24 Sep 2020 11:56:10 GMT

UPDATED 24th September 2020 with new data

Following on from our State of the Nation report in July, we look at four changes to society that we think are likely to become a way of life.

Slides can be downloaded as a PowerPoint deck at the end of the article. Please feel free to use them as you wish.

Working from home is here to stay

Lockdown has shown that many people do not need to go into the workplace to be productive. Working from home can be just as effective and has been embraced by over one quarter of the workforce.

ONS data for the period 24th August to 6th September 2020 shows that:

27% of the workforce were working remotely
57% were in their normal place of work
12% were on furlough leave

There are, however, differences by sector. Seven in 10 staff working in the Information and Communications sector are working remotely along with just under two-thirds in professional industries. Health and social care workers and those in the water and waste management sector are the most likely to be at their normal place of work while food services and arts and entertainments are more likely to remain furloughed.

The proportion working from home has fallen slightly over the last few weeks as schools and colleges have reopened and the push to get office workers back into the office has taken effect. However, with workers once more encouraged to work from home when they can, it is highly likely that remote working for at least some or all of the working week is here to stay for a significant number of people.

We have embraced online shopping like never before

In February 2020, internet sales represented 20% of all retail sales. By the end of August 2020 that figure had shot up to 28.1%. At the same time the value of all retail sales (the amount spent) was up 2.5% compared with February 2020.

Although the proportion of food sales online had more than doubled in that period, it still only represented 10% of all food sales.

Each category has seen significant shifts in purchase behaviour. Online sales in clothes & shoes now represents 29% of sales in the category (from 20.6%) and sales through department stores also represents 29% of all department store sales (up from 16.5% in February).

For clothing & shoes the shift towards online sales has not been sufficient to compensate for the lack of spend in store. The volume of sales in this category is down 16% on its February 2020 level but this is an improvement compared with the July figure of -26%. Department stores are slightly below pre-lockdown levels but household goods are above fuelled by purchases from DIY stores.

The volume of all retail sales made through the internet in August was actually down by 2.5% on the previous month and in July it was down 7% on June. It is, however, up 47% compared with February which suggests that buying online is a habit that will stick for many.

Consumers are looking for value and will be blending online with offline purchase journeys. Retailers will have to be value conscious (and by association cost conscious) but also need to ensure they have the logistics in place to make the blended experience seamless.

We are not visiting leisure, entertainment and retail attractions as much as we used to

Data from Google's Covid 19 Community Mobility Report (11 ^th September) shows that visits to restaurants, cafes, shopping centres, theme parks, museums, libraries and cinemas are down 16 percentage points compared with pre-lockdown February 2020.

The data also show that visits to public transport hubs (underground, Metro, train and bus stations) are down 28% compared with pre-lockdown days.

We are, however, making good use of parks and beaches to enjoy the summer weather. Visits to these places are up 97% compared with the February 2020 baseline.

We are visiting leisure, entertainment and retail attractions less at the back end of summer than we did during the middle of winter. Given recent spikes in Covid-19 infection rates, it seems that the number of people visiting attractions will not return to normal levels anytime soon.

This data is from mobile phone users who have enabled location data to be collected.

We are less confident, more anxious about the future

Consumer confidence is low. As at 17th September 2020 the Freemavens index of consumer confidence (known as Covindex) stood at -40.4 (+100 being the highest and -100 the lowest on the scale). Prior to lockdown the Covindex was in positive territory but only just.

Although there has been a slight recovery in the index since lockdown was imposed, the long term trend is one of anxiety about the future. This obviously relates to both the state of the economy and resultant pressures on the job market but also the possibility of local lockdowns or worse as Covid-19 infections accelerate as we head for winter.

In order to appeal to nervous consumers, brands will need to stay relevant, show empathy and deliver value in their offer (be it products or customer experience).

Freemavens Covindex is based upon an analysis of Google searches on around 400 variables. These include searches in relation to mental health and wellbeing, job security, financial worries and purchase intentions.

Although very different from survey-led confidence indices (GFK Consumer Confidence Index for example), Covindex provides a very similar direction of travel as can be seen from the latest GFK graphic below.

We believe that these four changes are likely to continue well into the future. The implications for each business sector will vary and are likely to be different according to each market.

We have views on their implications across different sectors and would be happy to share with you. Just get in touch for a chat .

Handy charts from this article for you to download and use as you like

Three things we've learnt from applying data analytics to market research

183:752711313 (Michael Wagstaff) — Fri, 04 Sep 2020 09:51:36 GMT

Human Made Machine Learning

1. There is no such thing as a mutant algorithm. An algorithm is just a set of instructions for a computer to follow. Yes, it's true that algorithms can learn off each other and evolve but the key issues are: who controls the algorithm and the parameters and assumptions baked into it; the source and quality of the data going into it.

The old saying ‘garbage in, garbage out’ is as relevant today as it’s always been.

2. Machines are not good at nuance and sarcasm. Machine learning helps make sense of big data but there are some things that cannot be easily learnt. It needs a human to make sense of the output.

3. Social media along with ratings and review sites represent a huge pool of voice of the customer data. There is so much consumer insight out there. It can tell a brand what consumers like about their product or experience, what they don't like and how it compares with rivals.

It works best when used in combination with expert human analysis.

That's why we call it Human Made Machine Learning.

We need to talk about Stanley

Wed, 12 Aug 2020 13:07:56 GMT

Dominic Cummings's trip to Barnard Castle has been blamed for undermining trust in government. Did Stanley Johnson's trip to Greece have the same effect and are there wider learnings for brands from an analysis of the Twitter storm?

Last month it emerged that Stanley Johnson, father of PM Boris Johnson, had gone to his holiday home in Greece. In normal times this would not be newsworthy but these are not normal times. At the point Stanley went to ‘Covid-19 proof’ his property, only essential international travel was allowed from the UK. Further, he went via Bulgaria to get round Greece’s ban on direct flights from the UK. Stanley's trip followed hot on the heels of Dominic Cummings's jaunt to Barnard Castle in County Durham to test his eyesight.

Predictably, Twitter went into meltdown. But was this just the ‘triggered’ venting their spleen at Johnson (Stanley, Boris, tick all that apply) or is there something more nuanced that a deeper analysis of the tweets could reveal? And what does the analysis tell us about the application of text analysis and machine learning in research for brands?

We analysed over 45,000 tweets using natural language processing and machine learning to undertake sentiment analysis and derive a list of topics discussed.

The analysis generated a wide range of topics but we were able to sort them into five main categories plus a long tail of ‘others’: The main categories are:

Undermining government advice (33% of all tweets). This is where most of the outrage happens. Tweets here reference travel restrictions both in the UK and Greece. The sentiment on Twitter was strong: Stanley’s actions undermines government guidance that only essential travel outside the UK was permitted. It also pointed out that Greece itself had restrictions on people entering from the UK and that Stanley had flown in from Bulgaria to get round these.

Abuse of power (21%). A common theme is the view that there is one rule for the elite and another for everyone else. Tweets relate to how it is fine for the elite to break the rules but everyone else is expected to follow them. Interestingly, there is a strong association made between the actions of Stanley Johnson and those of Dominic Cummings and his trip to Barnard Castle.

Descriptors (16%). These are tweets that set the scene and establish the events. Often these are tweets containing links to news reports and those establishing that Stanley Johnson went to Greece and that he is Boris Johnson’s father.

Message to the UK public (10%). This category of tweets asks what sort of message is it sending out to the public? Points were raised about the need for everyone to work together to contain Covid-19 by sticking to the guidance. People are giving up on their freedoms but the message seems to be that it is now OK to do what you want.

Brexit hypocrisy (4%). This category is about what is seen as hypocrisy among Brexiters like Nigel Farage and Nigel Lawson who have applied for EU passports for themselves or their families. Stanley Johnson has applied for a French passport and tweets were quick to label him a hypocrite despite it being well-known that he voted Remain in the EU referendum.

It’s quite clear then that the main take away is that Stanley Johnson’s actions were perceived to undermine government advice and were part of a pattern of behaviour of high profile figures not following the rules.

But the analysis also shows a great deal of concern among the public that ignoring safety advice might become the norm and this has profound implications for dealing with Covid-19 over the longer term.

What then are the lessons for the government on Twitter’s reaction to the Stanley Johnson event? The analysis highlights the importance of a clear message being consistently interpreted and universally applied. Actions that dilute the message risk undermining public trust in the message giver and ultimately the public’s compliance with it. We saw this with Dominic Cummings and we see it again with Stanley Johnson. Research published in the Lancet found that Cummings's trip to Barnard Castle may have reduced compliance with lockdown. The same is true of Stanley Johnson's trip to Greece.

This analysis has been about Stanley Johnson and the clear lessons around compliance. But the analysis also reveals some wider lessons for brands in the use of text analysis and machine learning. We have identified five that are relevant:

Social media along with ratings and review sites represent a huge pool of voice of the customer data. It can tell a brand what people like about their product or experience, what they don’t like and how it compares with rivals. It can identify what drives purchases and what puts people off. Why would a brand not wish to plug into this real time feedback?
People often get things wrong. We saw this quite clearly in the analysis as Stanley Johnson, a well-known Remainer, was labelled a hypocritical Brexiter. Text analysis needs someone who knows what they are doing to quality control the data.
Category damage can impact reputation. People are quick to make associations and sometimes these can work against a brand. Stanley Johnson and Dominic Cummings is an extreme example but a bad player in the category can tarnish the reputation of others. Monitoring your brand on social media is vital to managing your reputation.
Machines are not good at sarcasm and nuance. Machine learning helps make sense of big data but there are some things that cannot be easily learned. It needs a human.
Text analysis is very cost effective. Processing the insight contained in 45,000 tweets can be undertaken in a matter of hours rather than the days and sometimes weeks that most surveys take. The answers are already out there. Often there is no need for new primary research.

Text analysis is a really powerful tool. We used it to analyse tweets but it can be applied to any dataset that contains large quantities of unstructured text (such as customer experience data, ratings and review sites, blogs and so on). The insight it provides is maximised when used in combination with expert human decoding of the data.

That's why we call it Human Made Machine Learning.

Michael Wagstaff is a Partner and William Wagstaff Principal Programmer at SPARK Partnership.

Obesity and behaviour change : A personal view

183:752711313 (Michael Wagstaff) — Mon, 27 Jul 2020 19:06:44 GMT

The UK government has published its strategy on obesity. By applying a behaviour change model I plot my own journey from obesity to healthy weight

The UK government has published its strategy for tackling obesity. It includes a ban on junk food advertising before 9pm, the end of buy one get one free (bogof) offers on unhealthy foods, displaying calories on menus and a Better Health campaign to encourage people to eat less and be more active.

About one quarter of adults in the UK are obese with roughly another four in 10 overweight so it’s clear that a strategy is needed. But how do you get people to make a behaviour change that will not only start them on the journey to healthy weight but help them maintain it when they get there?

First off, I’m not going to summarise the literature on behaviour change and weight management interventions but if you are interested then the NICE website is a good place to start.

Instead, I’m going to draw upon my own experience and what motivated me. To do this I'm going to apply the Transtheoretical Model (TTM) of behaviour change to my situation. The TTM, developed in the 1970s by Prochaska and DiClemente, is widely used in market research and sets out a number of stages of change:

Pre-contemplation - no intention to do anything or do not recognise that action is needed
Contemplation - intend to do something in the foreseeable future
Preparation - taking steps towards doing it
Action - actually doing something about it
Maintenance - keeping the behaviour going
Termination - no intention to relapse into unhealthy behaviour

I was fat for about 20 years. I liked eating anything I wanted without restraint. At my heaviest (in August 2014) I weighed 18 ½ stones (117 kilos or 260 pounds). My Body Mass Index was 40 meaning that I was morbidly obese. Being fat was killing me. I felt ill most of the time, had no energy and was tired and irritable. I probably had type two diabetes or at the very least was at serious risk of it. Sugary foods and fizzy drinks were my medicine of choice. These gave me a short term fix but always left me craving more. Eating was an addiction for which I didn't want the cure.

On 1st September 2014 I decided enough was enough. Ten months later I had lost 7 stones (44 kilos or 98 pounds) to become healthy weight. What happened to make me go from Pre-contemplation to Action?

The scientific literature will tell you that personal motivators are hugely important in health behaviour change. Particularly important is the idea of autonomous motivation. This is when a change in behaviour is personally endorsed and engaged with. In other words, you do it through choice not because you think you should or ought to. The state cannot legislate or strategise for this. It can make bad behaviour harder (banning bogof) or good behaviour easier (listing meal calories in menus) but it can’t make people do things they don’t want to.

Moving to the Contemplation stage (intending to do something about it) and carrying this through to Action requires a personal motivation. Tom Watson, former Labour Party Deputy Leader, has said that wanting to be there in later life for his family motivated his dramatic weight loss. That’s also true for me. Others will have different motivations – it could be health, could be body image related, could be anything. The important thing is that getting to Preparation stage requires a personal motivation.

Having taken action, two things kept me going. The first was self-regulation. In the same way I made the decision to start weight loss I needed to make the decisions to implement the action and stick to it. Self-regulation for me was about having a daily calorie limit, eating more healthily, controlling portion size and doing exercise regularly. I stuck to the plan as if my life depended on it which it probably did.

The second important factor at this stage was setting goals and monitoring them. I set myself a target of losing 2lbs (one kilo) a week. That was my goal. I weighed myself weekly (frequent self-weighing is an important intervention) and charted my progress on an Excel file. I measured everything I could (I’m a quant researcher after all) – daily steps, weekly weight, distanced cycled and compared and contrasted against previous efforts.

Having a goal and frequent measurement of it is a really important weight loss intervention. I did this on my own but others like (or need) to be part of a group and have common goals and shared experiences. This is why weight loss classes can give people support to keep motivated.

What about Maintenance? What kept me going? The literature will tell you that having a more positive body image is a great motivator as is being determined to keep hold of the wins. I set myself another target. I read somewhere that only 1 in 1,200 formerly obese men stay healthy weight for five years. Most put the weight back on within 12 months of losing it. My motivation was that I wanted to be that 1 man in 1,200.

On 1st July 2020 I reached the five year healthy weight milestone. I celebrated with a bowl of trifle. I took a couple of spoonfuls, felt queasy and threw it away. Does that mean I’ve reached the final stage of the TTM known as Termination? This is when people have no desire to return to their unhealthy behaviour and are sure they will not relapse. The answer is: I don’t know. I hope so but for me losing weight was not about ‘going on a diet’ but about a complete lifestyle change that I know I have to adhere to for ever.

At this point I should say that my experience is just that: my own. I’m not saying that because I went from Pre-contemplation to Maintenance that everyone can. Far from it. There are a range of complex psychological and emotional issues around weight and body image and what worked for me won’t necessarily work for others. Some people celebrate their size and the body positive. That’s all good.

It should also be pointed out that deprivation and social inequality plays a significant part in health inequalities. The social, economic and environmental conditions in which people live have an impact on their health. So too do the commercial priorities of Big Food – deprived areas are much more likely to have fast food takeaways than non-deprived areas.

But for those who want to adopt a healthy lifestyle, government interventions will only take them so far. Losing weight is hard work and that is why much more education, nutritional advice plus tailored support and guidance is needed to help people through the stages to Action and beyond.

State of the nation

Mon, 06 Jul 2020 11:54:29 GMT

Lockdown measures are now being eased in all parts of the UK. Further, from 4th July pubs, cinemas, restaurants and museums in England will open their doors again.

Using official surveys and publicly available market research, we ask what sort of a nation is it that is now emerging from months of lockdown and how likely are we to see some form of normality returning?

A nation worried about work and the economy - some more so than others

We see a nation in which many people are not actively working and have acute money worries. The ONS Opinions and Lifestyle survey shows that as at 7th June 2020 , one third (33%) of employees had not done paid work the previous week due to being furloughed, self-isolating, a temporary business closure, caring for others or being off sick. For the self-employed, the survey estimates that 20% had not done work that week.

The economic impact has not been uniform across the population. People with low and middle incomes are the most adversely affected by the economic fallout of coronavirus. Those with a personal income of up to £20,000 are the most likely to have experienced a reduction in income and least likely to be able to save for the year ahead.

Economic well-being by income band, GB, May 2020, ONS

Conversely, those who earn over £40,000 are less likely to have been furloughed, experienced reduced working hours or a reduction in income and are most likely to have retained the ability to save money in the year ahead.

Although we see great concern among some income groups in relation to personal finances, generally speaking there is a disconnect between what people think will happen to the economy and the impact it will have on them. Populus’s Covid19 Public Perception tracker (June 30th) reveals a net worry about the economy score of 83% (% of people worried minus not worried). However, when it comes to worry about their finances the net worry score is 41% and for job security it is 38%. People think bad things are going to happen but the majority don’t think they’ll be affected by it. This could be true. But it could also mean an unfortunate economic surprise for a lot of people.

The disproportionate medical impact of the virus on Black and minority ethnic communities is well documented.

Research for the Fawcett Society shows how this affects anxiety levels among BAME communities. BAME women are particularly worried about being in more debt post Covid-19 than before – 43% compared with 37% of white women. For BAME men the figure is 35% compared with 34% for white men.

Similar differences were found in relation to whether people will struggle to make ends meet in the next three months. BAME men and BAME women were much more likely to be worried about paying their rent or mortgage compared with white men and women (38% BAME women, 37% BAME men compared with 32% white women and 28% white men). Nearly two-thirds of BAME women (65%) and just under three-quarters of BAME men (74%) feel anxious about going out to work during the pandemic.

An anxious nation suspicious of others

As we emerge from lockdown we find a nation showing high levels of anxiety and suspicious of other people.

MindMover’s Emotion Index has been tracking consumer feelings fortnightly since the start of the pandemic. General feelings among consumers are still very low although have improved slightly over the last few weeks. Feelings towards fellow citizens has improved over the last two weeks having previously plummeted as incidents of disregarding social distancing increased.

Emotion Index over the first seven waves,UK, June 30th 2020,MindMover

Sentiment towards the government’s handling of the crisis continues to be negative.

Are people more positive about the type of society they live in? YouGov polling in mid June found that two thirds of people thought that the UK was getting a worse place and only 5% thought it getting better. The question was very general and people will have answered it based on their individual circumstances rather than a common view but it does highlight a nation not at peace with itself.

With high levels of anxiety it is not surprising that huge numbers of people are having their sleep disrupted.

The excellent Rebecca Ironside’s Made You Think has been running a Covid-19 qualitative community and has some interesting insight on sleep. She found that dreams have become more vivid during lockdown and less restful and that people are actually going to bed earlier. Despite the earlier bed times, sleep is poorer than pre lockdown days.

People aren’t sleeping and they’re not having more sex! The fabulous Joe Twyman at Deltapoll has some insight into our sex lives . Deltapoll found that, overall, for half the population the amount of sex that they are having is the same as pre lockdown but there are quite significant variations according to whether you live on your own or with a partner. For one in five people living with a partner the lockdown has resulted in more sex but for two-thirds of single people it has resulted in less. On these figures, a baby boom in 2021 is not looking likely.

A nation prepared for a long haul back to normality

People do not expect a return to normal life anytime soon. The ONS Opinions and Lifestyle survey shows that the longer the pandemic has gone on the longer people think it will take to return to normality. In early April, about one in 10 people thought it might take a year to get to normal. By early June it is one in four – a quarter of the population.

An interesting think piece from Opinium [PDF download] asks people to think about their attitudes and behaviours post lockdown. It compares findings from April with those of May. It reveals that over half of consumers (58%) feel apprehensive about the future and 65% think it’s going to be ‘really weird’.

People's predicted attitudes and behaviours post-lockdown NET Agree, UK, May 2020, Opinium

Two months into lockdown and people were less likely to want to go abroad on their holiday. Probably the only surprising thing about this is that just over one third were even thinking about it. The proportion who thought they would go back to the old habits and behaviours also fell between the waves. Fieldwork was conducted in mid May so attitudes will have changed since then. Nevertheless, the data leads Opinium to conclude that normal isn’t coming back anytime soon.

Most important issues facing Britain today, top mentions,GB, June 2020, Ipsos Mori

The Issues Index also sees the increasing importance of the Black Lives Matter movement with concerns about race relations up 22 percentage points. We are likely to see this figure increase.

There is also evidence that people are getting a bit fed up of coronavirus in the news. Research by Ofcom found that 34% are trying to avoid news of the virus up from 22% twelve weeks ago. Under 35s in particular are looking to avoid news on covid-19.

Businesses believe in their own bouncebackability

A survey of business leaders by YouGov shines some light on their thinking. Nearly eight in 10 business leaders expect the economy to be worse in 12 months time but they expect their own company to be more resilient: 39% said their company would be doing worse in 12 months compared with how they were doing pre lockdown. But a quarter (26%) thought they would actually be doing better than they were pre lockdown with FS leaders the most optimistic at 37%. The same survey shows that two thirds of businesses are likely to make redundancies once the furlough scheme ends.

The latest instalment of the fortnightly Opinium-Cebr Business Distress tracker [PDF download] (wave 4, June 2020) suggests that the threat of insolvency has receded for many businesses. In the latest wave two thirds of a representative sample of businesses said that they were at no risk of insolvency from Covid-19 related disruption. This is up from 41% at the back end of April.

Retailers seem a lot more optimistic than previously. In the previous wave, 40% of retailers said that they were at moderate or high risk of insolvency due to Covid-19. Two weeks later, following the news of reopening, that figure had dropped to 18%.

A nation unlikely to Spend! Spend! Spend!

Are retailers right to be optimistic? Will consumers go on a spending spree now that lockdown is lifting? While it is true that shoppers have returned to the High Street since non-essential shops were allowed to open, in England at least, the level is nowhere near pre-lockdown levels. Springboard estimated that footfall across the UK for week commencing 15th June was 45% up on the previous week . This, however, is still 54% down on the same period last year. And more recent research from Ipsos Retail Performance shows that things have not improved that much. On Saturday July 4th, footfall was down 52% year on year.

Market research surveys also produce a mixed picture in relation to consumers' intentions to spend again: some will, some won’t.

According to BVA BDRC’s Covid-19 tracker from 16th June just under one third of people (32%) are likely to have a bit of a splurge to treat themselves or their family when life gets near normal again. The under 35s, families and those living in the midlands are the most likely to splurge while older people and those in Scotland, Wales and Northern Ireland least likely to spend out.

With shops reopening, many retailers will be hoping that consumers, deprived of retail therapy for so long, will start to splash the cash on the High Street. If survey results are to be believed then many might be disappointed .

According to YouGov fewer than half will go back to previous levels of spending on retail and four in 10 will actually spend less than they did before lockdown.

A nation that will shop differently

RED C Research has been undertaking ongoing qualitative research with families in the UK and other countries throughout the pandemic. In relation to UK food shopping habits , their research concludes that the lockdown has changed the way we shop with greater reliance upon deliveries and local grocers for fresh fruit and veg. Brand affinity has gone out the window as ranges have been trimmed and stock shortages experienced. The new way of shopping has meant that more time has gone into meal planning as once a week deliveries have replaced multiple visits to the supermarket.

RED C believe that the changes in food shopping are here to stay: online food shopping is easier and saves both time and money. There’s an added bonus too for families – the arrival of the shopping can create excitement among children especially if there are treats that they have been allowed to order.

Cambridge MR has looked back through their archive of food testing reports to see what lessons can be learnt from the 2008/9 recession in relation to food and drink. Back in 2009, buying British was considered an important recession beating behaviour and is likely to be so too this time, especially as we leave the EU. Further, families felt the effects of the recession hardest and this could well add further momentum to the permanent discount model of Aldi and Lidl. Finally, behaviour in 2009 was all about price, value and recycling. Post pandemic, health benefits are likely to have a bigger impact along with the use of sustainable materials and responsible food sourcing.

A nation longing to socialise and have fun but reluctant to do so too soon

The excellent BVA BDRC Covid-19 tracker provides a lot of insight for the leisure and entertainments sector.

The wave from 16th June shows that as well as missing family and friends, people miss going out to pubs and bars. Families in particular miss going on days out to visitor attractions. However, when asked how long they think it will be before they will visit an attraction, the average time is perceived as being 4.1 months. Two weeks later (fieldwork June 23rd-25th) this lead in time has decreased to 3.6 months leading BVA BDRC to conclude that we may be getting closer to 'normal' lead in times for days out.

The most recent wave (published 1st July) shows that 22% of people say they are likely to visit a restaurant by the end of July. This is up from 14% prior to the reopening of the sector. Intention to visit the cinema by the end of July remains flat at 6%. The figures still suggest a certain reluctance for people to take the plunge.

How strongly are Britons missing various activities, UK, BVA BDRC/Alligator, 16th June 2020

With over a quarter of Britons missing going to the pub a lot, did their reopening on 4th July lead to a huge rush of people going to the Dog and Duck? YouGov data suggests that 5% went to the pub the weekend of reopening. Men are more likely to have gone than women (7% versus 4%). A similar percentage went to get a haircut at the weekend although we don't know how much these two activities were related.

Going on days out often involves travel by public transport. The BVA tracker shows a slight recovery in intention to take a train and bus compared with the previous wave. Just over one in five (22%) will take a train by the end of August (up from 19%) and one in four (26%) will take a bus.

Research by Savanta published in Museums and Heritage Advisor , presents a mixed picture. On the one hand, 41% of a sample of the UK population say having a day out is one of the things they are most looking forward to post lockdown (this is second only to visiting family and friends). However, most people who used to take days out regularly (66%) want reassurance that it is safe to visit an attraction before doing so. Reduced capacity, security guards ensuring that social distancing is adhered to and free hand sanitisers are top of the list of things that would make attractions safe to visit.

What does this all mean?

Trying to make sense of life during a global pandemic is a fool's errand but here goes.

On the whole the nation is taking a pragmatic approach to life during lockdown and how we emerge from it. Although longing to do the things that we used to - see friends, going out, socialise, having fun - we are reconciled to the future being a bit weird and having to endure restrictions to normal life for the foreseeable future. Generally speaking, we want to make the best of things as they are and have, by and large, adapted to the demands of the situation. We see this particularly in relation to food: shopping has moved to online, meals have been planned better and brand loyalty has been replaced by brand availability.

Pragmatic it may be but it is also a nation ill at ease with itself and others. We don't completely trust the government or people around us. This is an incredibly important point because it will dictate the speed at which things return to normal. If people carry on perceiving others to be a risk because they flout social distancing rules, or don't trust government messages around safety, then they will be reluctant to go to bars and restaurants, visit the High Street or have days out - despite really wanting to do so.

For many of us, the 'bunker mentality' of the last few months will be hard to shake off while levels of anxiety, fear and mistrust stay as they are. For many, avoiding other people has become the norm. The effect of this on the economy will be profound. It will govern whether the on-coming recession is V shaped, U shaped or longer lasting. It could also result in the anxiety and worry that many feel becoming more and more engrained with huge consequences for mental health as a result.

Of course, the nation is not made up of one homogenous group. There are many segments in our society and the response of each of these will be different. Some sections of society want to live for today and will return to normal life much sooner than others - we see this already with media reports of packed beaches, increasing numbers of holiday bookings and people splashing the cash in retail therapy. We are likely to see a short-term spike in consumer spending as a result but whether this is sustained is open to doubt. There is also a worrying danger that many might have underestimated the impact of the recession on their own personal finances.

The data also shows how those already disadvantaged have been hardest hit by the economic impact of Covid-19. Lower income groups and BAME communities in particular have seen their economic outlook worsen.

The nation seems more divided than ever and eradicating inequality has to be the number one priority for the post Covid-19 nation.

I f you like what you've read and want more of the same, click here to sign up to our T3 e-bulletin. T3 is a weekly skip through survey results, statistics, insight and fancy facts that grabbed our attention, made us raise an eyebrow, nod wisely or made our blood boil.