Consumer psychology: why people can't tell you why they buy

Consumer psychology is the study of what actually drives a purchase, as opposed to the reason the buyer gives for it. The two are rarely the same, and the gap between them is where most of the useful information sits.

Ask someone why they bought the shoes and they will tell you about the arch support. They are not lying. They have simply been asked a question nobody can answer from the inside. The reason arrives after the decision: a person reaches for something and then assembles an account of why, and the account gets built out of the reasons it is respectable to hold. Price. Quality. Value. A recommendation from someone who runs.

This is the single most expensive fact in commercial research. A survey asking why do you buy this? returns clean, confident, internally consistent answers, and those answers describe a decision that was made somewhere else.

What consumer psychology studies that market research doesn't

Conventional market research is good at two things: measuring what people say, and measuring what people do. Both are worth knowing. Consumer psychology is concerned with a third thing — the distance between them, and what is living in that distance.

A person can describe their own behaviour perfectly accurately and still be wrong about its cause. They bought the expensive olive oil. That is true. They bought it because of the harvest date printed on the tin. That is the part that isn't.

Five mechanisms that produce the gap

Most categories run on some combination of these. They are not personality types — the same person uses all five in a single afternoon, on different shelves.

1. The cover story

The reason given is the reason that is acceptable to give. It is usually price, quality or health, because those three are never embarrassing. The real driver is left out because saying it out loud would cost something.

2. Projection

People are fluent readers of everyone but themselves. Nobody can say what their own trolley announces. Show them somebody else's and they will tell you in a second: careful, anxious, showing off, not really paying attention. What a person notices in someone else's basket is reliable data about their own categories.

3. Identity maintenance

Some purchases are bought before the behaviour they belong to. The running shoes come first and the running is meant to follow. The purchase is the commitment, and it works as a commitment whether or not the running ever happens.

4. Loss, not gain

People stay with a brand long past the point of preferring it, because leaving would mean the previous decade was just habit. Ask whether they would switch and they say of course, if something better came along. Nothing comes along, because nothing is being looked for.

5. Social signal

Every visible purchase says something to somebody, and the buyer is usually the last person able to name the audience. A teenager checking which trainers the older kids wear will deny the check flatly, and will not be lying — they have never had to put it into words, so they have no words for it.

How to research a decision nobody can narrate

The method follows from the problem. If the obstacle is that people cannot see their own reasons, then asking them harder does not help. Asking them differently does.

1. Ask about other people. "What do you worry about?" gets an edited answer. "What do people around you worry about?" gets an unedited one, because there is nothing to defend. This is social circle questioning, and it is the single highest-yield change most research teams can make.
2. Ask about behaviour, not motive. What did you buy last, what did you put back, what did you nearly buy. Behaviour is recalled. Motive is constructed.
3. Ask in three tenses. What was normal five years ago, what is normal now, what will be normal in five. Beliefs that people will not state directly show up clearly in the shape of the line between the three.
4. Test which questions people lie about before you field them. Sensitivity is predictable. Knowing which items in a questionnaire will be distorted, and in which direction, is cheaper than discovering it afterwards.
5. Treat the stated reason as evidence, not explanation. The cover story is data. It tells you what this person thinks is respectable to want, which is a useful thing to know about a market.

What this can and cannot tell you

Consumer psychology names patterns. It does not, by itself, tell you which pattern is moving volume in a given market at a given price against a specific competitor. Anyone who says otherwise is selling the interesting half of the job and leaving out the part that requires fieldwork.

A survey asking "why do you buy this?" gets five clean answers. What it won't get is the reason underneath any of them — not because people are hiding it, but because they cannot see it.

Questions people ask

What is consumer psychology?

Consumer psychology is the study of what actually drives a purchase, as opposed to the reason the buyer gives. It treats the stated reason as evidence about the person rather than as an explanation of the decision.

Is consumer psychology the same as consumer behaviour?

They overlap. Consumer behaviour describes what people do — the switching, the basket, the repeat purchase. Consumer psychology asks what is producing it. In practice the two are studied together, because behaviour without a mechanism is a pattern you cannot act on.

Why don't people know why they buy things?

Because the reason arrives after the decision. People reach for a product and then assemble an account of why. The account is built from reasons that are socially acceptable to hold, and the person believes it. Self-report is honest and still wrong.

What is the say–do gap?

The difference between what people report in research and what they do in a shop. It is not mainly caused by dishonesty — it is caused by people having no direct access to their own motives.

Can you fix it with a bigger sample?

No. A bigger sample measures the same wrong thing more precisely. The fix is in the question, not the base size.