Skip to content

Do We Know What We Like?


People are notoriously bad at explaining their own preferences. In one study researchers asked several women to choose their favorite pair of nylon stockings from a group of twelve. After they made their selections the scientists asked them to explain their choices. The women mentioned things like texture, feel, and color. All of the stockings, however, were identical. The women manufactured reasons for their choices, believing that they had conscious access to their preferences.

In other words: “That voice in your head spewing out eloquent reasons to do this or do that doesn’t actually know what’s going on, and it’s not particularly adept at getting you nearer to reality. Instead, it only cares about finding reasons that sound good, even if the reasons are actually irrelevant or false. (Put another way, we’re not being rational – we’re rationalizing.)”

Our ignorance of our wants and desires is well-established in psychology. Several years ago Timothy Wilson conducted one of the first studies to illustrated this. He asked female college students to pick their favorite posters from five options: a van Gogh, a Monet and three humorous cat posters. He divided them into two groups: The first (non-thinkers) was instructed to rate each poster on a scale from 1 to 9. The second (analyzers) answered questionnaires asking them to explain why they liked or disliked each of them. Finally, Wilson gave each subject her favorite poster to take home.

Wilson discovered that the preferences of the two groups were quite different. About 95 percent of the non-thinkers went with van Gogh or Monet. On the other hand, the analyzers went with the humorous cat poster about 50 percent of the time. The surprising results of the experiment showed themselves a few weeks later. In a series of follow-up interviews, Wilson found that the non-thinkers were much more satisfied with their posters. What explains this? One author says that, “the women who listened to their emotions ended up making much better decisions than the women who relied on their reasoning powers. The more people thought about which posters they wanted, the more misleading their thoughts became. Self-analysis resulted in less self-awareness.”

Wilson found similar results with an experiment involving jams. And other researchers, including Ap Dijksterhuis of Radboud University in the Netherlands, have also demonstrated that we know if we like something, but we don’t know why and the more time we spend deliberating the worse off we are. Freud, then, was right: we’re not even the masters of our own house.

Our tendency to make up reasons for our preferences is of particular importance for advertisers, who sometimes rely on focus groups. But if we don’t know what we like, then how are ad agencies supposed to know what we like? The TV shows The Mary Tyler Moore Show and Seinfeld, for example, are famous for testing terribly even though they went on to be two of the most popular shows in the history of TV. By the same token, many shows that tested well, flopped. As Philip Graves, author of Consumer.ology reminds us: “As long as we protect the illusion that we ourselves are primarily conscious agents, we pander to the belief that we can ask people what they think and trust what we hear in response. After all, we like to tell ourselves we know why we do what we do, so everyone else must be capable of doing the same, mustn’t they?”

Stories of the failures of market research are not uncommon. Here’s one from

At the beginning of the ’80s, I was a product manager at General Electric, which at the time had a leading market share in the personal audio industry (radios, clock radios, cassette recorders, etc.). Sony had just introduced the Walkman, and we were trying to figure out how to react. Given the management structure of the day, we needed to prove the business case. Of course, we did focus groups!

Well, the groups we did were totally negative. This was after the Walkman had been on the scenes for months, maybe a year. The groups we did felt that personal music would never take off. Would drivers have accidents? Would bicycle riders get hit by drivers?

If we listened to “typical” consumers, the whole concept was DOA.

This type of reaction is probably the reason that there is the feeling of a “technological determination” on the part of the electronics community. It leads to the feeling that you should NEVER listen to the consumer, and just go about introducing whatever CAN be produced.

At the time, we had a joke about Japanese (Sony/Panasonic/JVC) market research. “Just introduce something. If it sells, make more of it.” It’s one way of doing business. One the other hand, when I was hired by a Japanese company in the mid-80’s, I was asked how GE could get by with introducing such a limited number of models. Simple, I said, “We tested them before we introduced them.”

History tells which method has worked better.

One person who understood this was Steve Jobs. He never cared for market research or focus groups because, as he once said, “people don’t know what they want until you show it to them.” Instead, Jobs was a pseudo- Platonist about his products. He believed that there was an ideal music player, phone, tablet and computer and trusted the customers to naturally recognize perfection when they saw it. When asked what market research went into the iPad, his New York Times obituary reports, Mr. Jobs replied: “None. It’s not the consumers’ job to know what they want.”

I’m not the only we with an ancient Greek take on Jobs. Technology-theory contrarian Evgeny Morozov compared Jobs to Plato a few years back. He said:

The notion of essence as invoked by Jobs and Ive [the top Apple designer] is more interesting and significant—more intellectually ambitious—because it is linked to the ideal of purity. No matter how trivial the object, there is nothing trivial about the pursuit of perfection. On closer analysis, the testimonies of both Jobs and Ive suggest that they did see essences existing independently of the designer—a position that is hard for a modern secular mind to accept, because it is, if not religious, then, as I say, startlingly Platonic.

Does this mean all markers should think platonically? Not necessarily; Jobs, to be sure, was an outliner. But it does remind us that many times we don’t know what we like. 

  • This is an update version of the post, which I originally published late Sunday night. Special shout out to the Farnam Street blog and the WSJ Ideas Market for some added quotes.


15 Comments Post a comment
  1. Jim Purdy #

    I like this post.

    I mean, I think I like this post. Oh, heck, now you’ve got me confused.

    To paraphrase Descartes, “I think, therefore I am … confused.”

    February 27, 2012
    • sammcnerney #

      Haha! Well, I am glad you liked it, whatever that may mean.

      March 1, 2012
  2. Reblogged this on the science of science and innovation: field notes and commented:
    These are interesting insights about people’s access to their preferences. I am skeptical of how well these ideas generalize to domains in which people are “expert” or in which the properties that map onto value/worth/preferences are transparently analyzable (e.g., tech equipment?). But a major point stands re: the skepticism with which we should view “mere” customer surveys when looking for or evaluating design opportunities. The wisdom of the recent insights of innovators re: the power of observation and ethnographic research comes into sharp relief in light of these findings about the psychology of preferences.

    February 27, 2012
    • sammcnerney #

      Hi Chozen86,

      Thanks for the repost. I think I might be skeptical too, but I would have to keep thinking about it.

      March 1, 2012
  3. Mark Wood #

    Often with consumer products I think many people have to be coerced by ads or pressured by culture into wanting stuff that they don’t wouldn’t want in the first place. Or maybe they just want whatever improves their perceived status in society and they differ to ‘the experts’ to tell them what that is.

    February 29, 2012
    • sammcnerney #

      I agree. I think that is right a lot of the time.

      March 1, 2012
  4. Tom Fairhurst #

    Interesting article. It does make you want to ask the question as to whether or not it is the research methodlogy that may be flawed or the questions being asked or stimulus being used to solicit the responses. At the end of the day, don’t we really want to understand what will change behavior and not necessarily what someone prefers or doesn’t prefer. What people like or don’t like (color preferences for instance) don’t alone change behaviors. The goal is to understand what does create disruption, why and why are you (the consumer or shopper) making the decision you are making. In addition to understanding the right questions, focus groups also require a different type of listening skill. It is listening to understand across multiple groups to discern threads of consistency or inconsistency. Not trying to defend the research methodology, but propose that personal preferences are only a piece of understanding behavior and, therefore, changing behavior to achieve a different outcome requires a much more in depth understanding with different questions.

    February 29, 2012
    • sammcnerney #

      I appreciate your skepticism. It’s also good to question the methodology.

      March 1, 2012
  5. Isn’t it more likely that, under the prospect of having to reveal something about themselves, the “non-thinkers” simply chose the item that would lead to a more superficial explanation? Someone who may have preferred Van Gogh (which can evoke or reveal certain emotions) simply skirts the whole problem by “choosing” something cute? They trade off what they really wanted to stay socially “safe.”??

    February 29, 2012
    • sammcnerney #

      …. Yes, that is possible. But I would guess only in a few cases.

      March 1, 2012
  6. Great post

    As a Market Researcher I’m constantly trying to get to what people meant, rather than what they said.

    Non verbal reactions like leaning forward/ smiling are really important. If you’re moderating a coupla times a week you can soon scent when someone is giving you the ‘yeahs’ (i.e. empty praise).

    The best solution? Triangulate Attitudes (e.g. Focus Group), Behaviour (e.g. sales data, customer database) and Context (e.g. norms, market price points). 

    My approach is summarised here



    March 8, 2012
  7. Wonderful post! We will be linking to this great post on our site.

    Keep up the good writing.

    October 18, 2013
  8. Para permitir el traslado y su posterior venta en España, en sus documentos constaba que todos los animales habían nacido hace 4 meses.

    October 3, 2014
  9. “If I had asked people what they wanted, they would have said faster horses” – Henry Ford

    January 5, 2015
  10. dpennock #

    Interesting article. Some typos: “only we”, “markers”, “outliner”

    January 6, 2015

Leave a Reply, Be Constructive, and Let's Debate

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: