The Free Sample Trap: Why On-the-Spot Feedback Tells You What People Think You Want to Hear

July 20, 2026

Six years ago we created a comic strip about a couple trying a free product sample at a shopping centre. At the stand, they smile and say “very nice, very tasty.” The sampler is delighted and sends them home with extras. Back at home, the truth comes out: “That was awful. Who would want to buy that?” “It was disgusting. I was just trying to be polite.”

We made it as a light-hearted illustration of a serious problem. Six years on, nothing has changed. Free sampling with on-the-spot feedback remains one of the most widely used methods for gathering consumer reactions, especially among SMEs launching new food, drink, fragrance and personal care products. It feels cheap, fast and direct. Unfortunately, it is also one of the least reliable forms of research a brand can run.

As Margaret Mead put it, what people say, what people do, and what they say they do are entirely different things.

A real-world lesson: New Coke

The most famous example in business history comes from a company with almost unlimited research budgets. Before launching New Coke in 1985, Coca-Cola conducted taste tests with close to 200,000 consumers. In sip tests, people consistently said they preferred the new, sweeter formula. The company had the data, and the data said go.

The launch triggered a consumer backlash so severe that the original formula was reinstated within 79 days. The research had measured what people said in an artificial tasting situation, not what they felt, chose and bought in real life. A quick sip in a test setting favours sweeter products, and stated preference in front of a researcher tells you very little about repeat purchase at home.

If it can happen to Coca-Cola with 200,000 respondents, it can certainly happen to an SME sampling at a food fair with 50.

Why on-the-spot feedback misleads

When someone hands you a free sample and asks what you think, several well-documented biases activate at once:

Courtesy bias. Most people simply do not want to criticise a product to the face of the person offering it. Saying “very nice” costs nothing and avoids an awkward moment. The comic captures this perfectly: the couple even took extra samples home just to be polite.

Reciprocity. A free sample is a gift, and gifts create a sense of obligation. Consumers repay the gesture with kind words, not honest evaluation.

Social desirability bias. People give the answer they believe makes them look agreeable and reasonable, particularly when other shoppers are within earshot.

The halo effect. The sample is rarely tasted blind. The branding, the packaging, the stand, even the friendliness of the person sampling all colour the response before the product touches the tongue.

First-taste bias. A single sip or bite in a busy environment is nothing like real consumption. Sweeter, saltier and more intense products win first tastes but often lose over a full serving, exactly what caught out New Coke.

Observation effects. People behave differently when they know they are being watched and evaluated. A shopping centre concourse is about as far from natural consumption as it gets.

The result is a wall of polite positivity. Around 40 percent of new products fail, and a major reason is that when consumers are asked directly, researchers have no way of knowing whether they are telling the truth. Brands walk away from sampling days convinced they have a winner, then watch the product struggle at launch.

What proper sensory testing looks like

None of this means consumer feedback is worthless. It means the method matters. Proper sensory testing removes the pressures that produce polite answers and replaces opinion with measurement. That means objective measures: a combination of implicit response testing, which captures instinctive reactions in milliseconds before conscious editing kicks in, and biometric measures such as facial expression analysis, portable EEG, which records brain activity as the product is experienced, and physiological arousal captured through medical-grade wearables such as Empatica. These devices record how the brain and body actually respond to a product, and these responses cannot be faked to be polite.

It also means genuinely blind testing. Products are evaluated without brand cues, so you learn which flavour, fragrance, formulation or texture truly performs, not which logo people recognise. And it is not limited to flavour. The texture and mouthfeel of an early prototype can be tested the same way, long before you commit to final production, because texture is often where a product wins or loses on the second and third use, well after the first polite taste. When you test three flavour or texture variants this way, you do not get three sets of compliments. You get a clear, honest ranking of which one consumers instinctively want, and which one would quietly die on the shelf.

Now within reach of SMEs

For years, this kind of research was only accessible to the biggest corporations, the companies that could afford dedicated sensory labs and six-figure research budgets. SMEs were left with clipboard surveys and sampling tables, the methods most vulnerable to the biases above.

That has changed. New technology has made implicit and biometric sensory testing dramatically more affordable and accessible. Portable EEG headsets have replaced room-sized lab equipment, and medical-grade wearables such as Empatica can capture physiological responses reliably in real-world settings, in a kitchen or a tasting session rather than a laboratory. Combined with Deeplight, our behavioural research platform, an SME can now run the same class of objective, blind sensory research that global brands use, at a fraction of the traditional cost and timeline.
Before you commit to production runs, packaging and a launch, you can know with confidence which flavour, texture or prototype consumers would definitely want. Not which one they said was “very tasty” to be polite.

The couple in our comic told the sampler exactly what she wanted to hear, then binned the product at home. Six years later, consumers have not changed. But the tools available to you have.

AT THE SHOP

“What people say, what people do, and what they say they do are entirely different things”
– Margaret Mead

AT HOME

How they truly feel

40% of new product fails because when consumers are asked about the product they say they would buy it, but researchers do not know if they are telling the truth.

Split Second Research’s online test of the subconscious can’t be faked, so brands can measure how consumers truly feel.

Visit our website to find out.

References

The Coca-Cola Company. New Coke: The Most Memorable Marketing Blunder Ever? https://www.coca-colacompany.com/about-us/history/new-coke-the-most-memorable-marketing-blunder-ever

Encyclopaedia Britannica. The Not-So-Sweet Backlash to New Coke. https://www.britannica.com/today-in-history/April-23-The-Not-So-Sweet-Backlash-to-New-Coke

Castellion, G. and Markham, S.K. (2013). Perspective: New Product Failure Rates: Influence of Argumentum ad Populum and Self-Interest. Journal of Product Innovation Management, 30(5), 976-979. https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1540-5885.2012.01009.x

The Behavioral Scientist. Courtesy Bias. https://www.thebehavioralscientist.com/glossary/courtesy-bias

Scribbr. What is Social Desirability Bias? https://www.scribbr.com/research-bias/social-desirability-bias/

Empatica. EmbracePlus medical-grade wearable. https://www.empatica.com/embraceplus/
Split Second Research. https://www.splitsecondresearch.co.uk

Latest Posts