Why Consumers Aren’t Good At Telling You What They Want

When Steve Jobs was asked if Apple had used consumer research to help design and launch the iPad, he replied “No.” And when asked why not, he said, “It’s not the consumer’s job to know what they want.”

While many us might not want, or might not be well advised to trust our guts to the extent that Jobs did, we could learn from his skepticism of asking people what they think and feel and taking their answers as definitive information to drive our marketing to them.

Actually, a better answer for Jobs to give would have been “The consumer isn’t very good at telling you what they want.” Yet most marketing research is based on asking them what they think about products, brands or ideas, whether it does this by questioning them in focus groups or having them answer direct questions in quantitative research.

There are a number of reasons why people aren’t good at telling you what they want, or what may influence them, especially in traditional research situations.

1. Research makes people pay more attention than they do in the real world.

Robert Heath, University of Bath Management School and author of “The Hidden Power of Advertising” has written a lot about the theory that we tend to be in one of two modes when we are processing information. One is High Involvement Processing, which is when we are actively paying attention to something, the other Low Involvement processing, which is when our attention is running in the background. These modes seem to work in quite different ways; high involvement processing enables us to remember logical detail and recall limited amounts of it very accurately, but often only for a short period of time after paying attention. These memories are triggered voluntarily. Low Involvement Processing, on the other hand, seems to seed memories that are triggered by external episodes where there is some association. These are often more emotional, last longer and seem more powerful in terms of evoking action. The evolutionary psychology theory might be that Low Involvement Processing is always running in the background when we are focusing on other tasks. It would have been our way of picking up information in the broader context that helped us form an overall and connected view of our environment, to help us survive and thrive over the long term. As we ambled out foraging we might have subconsciously picked up a sense of the environmental factors that point us towards successful hunting and gathering. High Involvement Processing is more about memory relating to specific tasks. For example, if we found ourselves exploring new territory, we might deliberately have tried to remember landmarks to guide us home.

We have seen aspects of this through the work of Christopher Chabris, who has been one of our advisors at the Institute of Decision Making. Chris conducted the very famous “Invisible Gorilla” research while at Harvard, and subsequently published a book with the same name. Without spoiling the book or experiment, his research shows that if people are primed to pay attention to a piece of film, and given a memory task, they will recall detail but have a very high chance (up to 50%) of missing non-related events that may be more significant and more emotionally loaded. We seem to pick up emotions better when we aren’t actively paying attention.

The two major things Robert Heath points out that are important to marketers are:

(i) Nearly all consumer research puts respondents in a high involvement processing mode and gives them high involvement processing tasks like detail recall, while the nature of much of the media exposure that our brands get is likely to be consumed via low involvement processing.
(ii) For many brands, entering memory through Low Involvement Processing may be more helpful for marketers, as it allows the brand to forge emotional connections, and external associations can trigger these memories at a later date.

2. We’re not as rational as we think

Gerald Zaltman Joseph C. Wilson Professor Emeritus at Harvard Business School and the author of How Customers Think (2003) and Marketing Metaphoria (2008) believes that 95% of decision-making is unconscious; Gary Klein, whose research pioneered the field of Naturalistic Decision Making, and whose findings led to significant changes in military training, estimates that 90% of the critical decisions we make is based on our intuition; and Daniel Kahneman tells us our intuitive system “is more influential than your experience tells you, and is the secret author of many of the choices and judgments you make”.

Dan Ariely (who has been a great friend to the Institute) dissects this in his excellent books “Predictably Irrational” and “The Upside of Irrationality.” We hate to see ourselves as irrational, which is why respondents in research will often choose a more rational approach or one which is easy to rationalise when asked to make a choice. But the fact is that the reasons we attribute to decisions that we’ve made are very often not the reasons at all. Most rational reasons are post-rationalizations of emotional or instinctual decision making processes, which were not understood even by experts 50 years ago, and which have their roots in evolutionary psychology. Even people who are aware of the fundamentals of Behavioral Economics often find it hard to explain their own behavior in the heat of the moment. Yet we spend billions of dollars asking respondents in our research to do exactly that.

3. If people think they are going to have to explain a choice, it affects the choice they make

In a famous study by Timothy Wilson and Douglas Lisle at the University of Virginia, two cells of respondents were asked to choose a poster from a range that went from representative (let’s say a photograph of two puppies playing) to abstract (modern art). Both cells were told that they could return their poster if they didn’t like it. But before they made their choice, one cell was told that they would be asked to explain why they liked the poster they chose. Two very interesting things happened. The first was that there was a notable difference in the types of posters chosen by each cell. The cell that didn’t have to give their reasons chose, on average, more abstract posters. The cell that did have to give their reasons chose posters that were more representative. The need to explain leads us to make choices that we can explain – two puppies can be explained as “I used to have dogs as a kid,” whereas a preference for print of a Modigliani is more difficult to put into words.

The really interesting finding was when the respondents were contacted three weeks later to see if they were happy with their posters and whether they wanted to change them, the respondents who had been asked to give their reasons were significantly less satisfied with their choice. As the abstract of the research states:

“When people think about reasons, they appear to focus on attributes of the stimulus that are easy to verbalize and seem like plausible reasons but may not be important causes of their initial evaluations. When these attributes imply a new evaluation of the stimulus, people change their attitudes and base their choices on these new attitudes. Over time, however, people’s initial evaluation of the stimulus seems to return, and they come to regret choices based on the new attitudes.”

So, it’s not just that people can’t tell you why they might make a certain choice, if they know that you are going to ask them why, they may make a different choice, and that choice may be one that they will be less happy with!

4. What people say they like can change.

What people say in a research setting is dictated by their mental and physical state at the time. Neuroscience shows that when people are hungry, they respond differently to images of food than when they aren’t. A bland research room with office furniture may be more stressful than your couch at home, but less stressful than the weekend shop. Behavioural research shows our levels of stress affect how we respond to information.

But it doesn’t stop there. Sheena Iyengar, author of ‘The Art of Choosing’ describes an experiment where people were shown pictures of two attractive women – one blonde, one brunette.

“… (participants) were shown a whole set of different pairs of female pictures, and asked which ones they thought were prettier. They were then shown the pictures they chose again and asked why they picked them. In some cases though, unbeknownst to the respondent, the images were switched – if they had chosen the brunette they were sometimes shown the blonde. What did they do? 87 per cent of the time they didn’t even notice. They simply said, “Oh, I prefer blondes”… even though they had actually chosen the brunette!”

All of this is to really say that we now know enough to know that asking people to reveal how they make decisions, and what may influence them isn’t going to help us make the best decisions as to how to market to them.

What might lead us to better research, then?

The prime purpose of the Institute of Decision Making is to stay at the forefront of discoveries and emerging thinking in the field of decision sciences. In doing this we are becoming increasingly knowledgeable in three areas:

1. Cognitive biases and heuristics that drive the unconscious aspects of decision-making. Much of this falls under the area of Behavioral Economics. We have worked with some respected behavioral economists and have connections with many others.  The key here is to use a sound understanding of these principles to diagnose the underpinnings of existing consumer behavior, and what these principles suggest could lead to changes in that behavior.

2. Neuroeconomics, which is the application of neuroscience and other biometric measures to understand cognitive load and arousal whether as reactions to stimuli, or during a choice test. We have become reasonably fluent with the main methods, and have an advisor at Stanford who helps us understand the real science rather than taking the sales pitches at face value, and we were one of the groups that ARF sought input form for their 2010 review of neuroscientific approaches to measure advertising. This is a field where many wild claims are made; we remain convinced there is value here but it may be at its greatest in areas outside of neuro-copytesting.

3. Implicit Association. This covers a range of techniques where the time in which it takes to respond is used to understand the real feelings of the respondent about stimulus. A straightforward implicit association test shows the strength with which people agree with or feel affinity towards something from the speed with which they respond (faster meaning stronger or more instinctual), while a variation called a Stroop test shows when a thought causes dissonance or requires an inhibited response, resulting in slower responses.

None of these techniques is a silver bullet, and there is some evidence that many work best in conjunction with more traditional research. But all of them go beyond asking people what they think and taking those answers at face value. Our aim is to keep pushing the envelope in understanding what is new and interesting in these areas, so that we can help marketers understand what consumers can’t tell them. Not because it’s not their job, but because they are poorly equipped to do so.