The say-do gap: One of the biggest challenges in shopper research
Editor’s note: Patrick Young is the managing director and a global board member at PRS IN VIVO, U.K., with over 15 years of experience in qualitative and quantitative research projects. He holds a psychology degree from University of Bristol and has a certification on behavioral science for business success. Find Young on LinkedIn.
Around 70-90% of new fast-moving consumer goods (FMCG) products fail, despite many having passed extensive concept testing, consumer surveys and purchase intent measures before launch. It raises an uncomfortable question for our industry. If we spend millions asking consumers what they will buy, why do so many products still fail when they reach the shelf?
The answer may not lie in asking better survey questions, but rather in recognizing that much of shopping happens before shoppers are consciously aware of the decisions they are making.
For decades, shopper research has relied heavily on self-report. We ask shoppers what they noticed, what influenced them and why they chose one product over another. These answers are invaluable for understanding attitudes, motivations and beliefs. Behavioral science reminds us that they often capture only part of the story. The problem isn't that shoppers are misleading us, it's that we are asking them to explain decisions that were never fully conscious in the first place.
Shopping is designed for speed, not reflection
The modern supermarket is one of the most cognitively demanding environments we encounter. Tens of thousands of products compete for attention, with many categories containing well over one hundred alternatives. Yet shoppers don't spend minutes evaluating every option. Our Retail Lab research shows that category decisions are often made in around 2.2 seconds, with shoppers ignoring approximately 80% of the products available on shelf. Mobile eye-tracking further suggests they consciously process only around one quarter of the SKUs in front of them. This is efficiency at work.
If we consciously evaluated every product on every shopping trip, a weekly grocery shop would take hours. Instead, the brain simplifies the task through habit, visual shortcuts and rapid pattern recognition. Daniel Kahneman described these as System 1 processes: fast, intuitive and largely automatic. While researchers often question shoppers as though purchasing were a slow, analytical exercise, the reality is that much of it is driven by subconscious filtering long before deliberate reasoning begins.
The say-do gap isn't a flaw in respondents
The phrase "say-do gap" is often framed as though consumers are unreliable witnesses. I think that's the wrong interpretation; people generally provide honest explanations based on how they remember the decision: the challenge is that memory reconstructs behavior rather than replaying it.
Imagine a shopper tells us they bought a product because it "looked healthier." Traditional research might accept that explanation at face value. Whereas behavioral observation often reveals something more nuanced: Eye-tracking may show the shopper never looked at the nutrition panel, calorie count or ingredient list. Instead, they responded almost instantly to green packaging, natural imagery, simple front-of-pack claims and familiar brand assets. Health genuinely influenced the purchase, but not through conscious evaluation of nutritional information.
Both findings are true: The survey explains the shopper's conscious interpretation of the decision. Observation explains how that decision unfolded. Understanding the difference fundamentally changes where brands invest their effort.
Attention is the first purchase decision
Perhaps the most important lesson behavioral research has taught us is that shoppers cannot evaluate products they never notice. It sounds obvious, yet many research programs still begin by asking consumers which design they prefer, rather than whether they actually saw it. Behavior follows a sequence. A product must first attract attention before it can communicate its proposition, justify its price or persuade someone to buy it.
Eye-tracking illustrates just how demanding this challenge is. The average fixation on a product at shelf lasts only around 300 milliseconds before the brain decides whether it deserves further attention. In those fractions of a second, packaging must interrupt habit, signal category relevance, reinforce the brand and communicate enough meaning to encourage a closer look.
If it fails at the first hurdle, none of the carefully crafted claims, brand credentials or product benefits matter because they are never processed.
Observation changes the questions we ask
This is where behavioral methodologies become so valuable. Rather than simply asking whether shoppers liked an idea, they reveal where behavior succeeds or breaks down.
In our research, we've observed products that perform strongly in survey-based concept testing yet struggle at shelf because they simply disappear visually within the category. Equally, we've seen packs that generate exceptional stopping power but fail to convert because shoppers misunderstand the proposition once they pick them up.
Behavioral observation doesn't replace traditional research; it complements it by measuring the decisions that happen before shoppers are even aware they're making them. Those early moments have a disproportionate impact on commercial success. If a pack isn't seen, it can't build brand recall. If its primary benefit isn't understood, it can't drive purchase.
Our research consistently finds that visibility is one of the strongest predictors of brand recall, while benefit communication is strongly correlated with purchase. We also know that in-store remains the single biggest source of product discovery, influencing around three times more purchase decisions than all forms of advertising combined.
Behavioral measures provide an early indication of whether a product is likely to work in the real world before sales data tells you it hasn't.
Stop asking the end of the story first
Too often we begin by asking questions at the end of the decision journey.
- Which design do you prefer?
- Would you buy this?
- Why did you choose it?
Behavioral research suggests we should first understand what happened before those questions became relevant.
- Did shoppers notice the product?
- Did they recognize the brand?
- Did they understand the proposition?
- Did they hesitate?
- Did they compare alternatives?
Only once those behavioral stages are understood does it make sense to ask shoppers how they interpreted the experience. In other words, our research design should mirror the sequence of human decision-making rather than assuming every purchase begins with conscious evaluation.
The future isn't behavioral vs. declarative
None of this diminishes the value of surveys. Surveys remain one of the best tools we have for understanding conscious beliefs, motivations and emotional responses. Observation tells us what happened; surveys help explain what shoppers believe happened and how they rationalize those behaviors afterwards. As such, the strongest research programs combine both. Given around 60-70% of FMCG volume comes from repeat purchase, understanding both what creates initial trial and what builds long-term loyalty becomes critical. Observation helps explain the first, declarative research helps explain the second.
This distinction will become even more important as artificial intelligence transforms our industry. AI can analyze millions of survey responses almost instantly, but it cannot compensate for asking the wrong questions. If the input consists solely of conscious explanations for largely unconscious behavior, we simply become faster at analyzing incomplete answers.
The future of shopper insight therefore lies not in choosing between behavioral and declarative methods, but in recognizing that each illuminates a different part of the purchase journey. When we stop asking shoppers to explain decisions they never consciously knew they had made, we move much closer to understanding what really happens at the shelf.