Writing a new epilogue for the paperback affords me the opportunity to answer some of the questions people have raised after reading the first edition.
I accept that the term “market research” encompasses a very wide and diverse set of techniques and yes, my primary grievance is with those techniques that involve asking people to explain what they think. However, many of the other techniques would benefit from the same considerations of psychological validity that I advocate here. Are people who are taking part in an ethnographic study, and who know they are being watched, going to behave “normally”? They may or they may not; the mistake is to presume that they will. Even techniques like eye tracking involve significant alterations to the “normal” consumer context. I’m not saying that these techniques can’t work, only that care should be taken to check that the behavioral outcome is the same as from a group that aren’t being told to put on glasses and go shopping.
Certainly not. People can generally tell you what they are doing right now. They can even tell you, reasonably reliably, how they feel at that instant in time. A few moments either side of now and you’ll still get a fairly accurate response. Travel much further in time, though, and things become less reliable. Add in the influence of the question being asked and the, usually, unavoidable desire to ask other questions, and you get the problems of priming, framing, focalism, and so on.
People are often surprised to discover that I still undertake market research of the “asking questions” variety. The difference is that I start from the basis that people can’t reference the unconscious mental processes that determine their behavior. I have to use techniques that focus respondents on their behavior and leave me to decode the psychological drivers behind their actions from the clues I can pick up along the way. Doing so isn’t perfect, it can’t explore everything, it doesn’t help predict the future, but it can explain why people are currently doing what they do, and that’s a good place to start if you’re interested in changing it.
Questionnaires still have a place when live tests aren’t practical, but they need to be handled with far greater care and the results treated with far greater circumspection than is usually the case.
There is also the opportunity to conduct implicit testing: by measuring the speed of response to carefully designed questions it is possible to identify someone’s unconscious (implicit) associations with a stimulus. This can provide an extremely useful insight into one piece of the puzzle.
Pretty much perfectly. Behavioral economics is the marriage of economics and psychology. Psychologists are perhaps a little more interested in understanding why people have done something, and economists in predicting what they will do, but both exist as sciences that seek to develop hypotheses and test them. I’d also argue that we operate on a scientific basis that is open to changing its approach on the basis of new evidence (not something that is always the case with market research).
I work closely with economists as an associate of Frontier Economics, and a number of academic behavioral economists have been very complimentary about the ideas in Consumer.ology.
At the moment there is a mostly unspoken tussle between the old way, asking people what they think, and the new behavioral methods. In part this is because we are so wedded to the idea that we know what we think and why we think it that what customers say feels like it must be of importance and value.
A number of people have suggested the “trigonometry argument” to resolve this tussle: they propose that more data is always better, and that what people say is something that should be factored in as one part of the puzzle when making a decision. The theory goes that by balancing up the information acquired from the various sources an optimum picture or direction will emerge. The flaw in this argument is, to continue the trigonometry analogy, that if one of your data points is bogus, referencing it simply pulls you away from a direction that fewer, valid data points would have helped you find.
This was neatly demonstrated by the behavioral economist Dr. Robert Metcalfe, who explored the impact of different messages about energy saving on people’s energy consumption. Initial market research, which relied on asking people what they thought, had concluded that telling people they could save around £50 by turning their thermostat down by one degree would bring about the desired outcome. However, when field experiments were conducted that involved people receiving one of a number of different messages intended to reduce energy use, and tracking those homeowners’ meter readings, it transpired that people receiving this message actually used more energy, not less.
This experiment, and many of the others conducted under appropriate experimental conditions, make one thing abundantly clear: it is really hard to predict what people will do in any given context. Asking other people or ourselves what seems reasonable is of little benefit, because our conscious introspection doesn’t connect with the processes that drive our behavior. Good experiments are required if we want to have any confidence in what the impact of an initiative will be.
Neuroscience is in its infancy and is still a long way from being able to get to grips with the complexity of the mind. However, a clearer picture of how we think is starting to emerge from the many experiments that psychologists and behavioral economists have conducted. This provides, I believe, a framework that dramatically increases the likelihood of understanding existing consumer behavior properly and of targeting the right way of bringing about change. At the very least, it helps ensure that all the possibilities will be considered when attempting to understand current consumer behavior or explore what might change it in the future.
At one end of the spectrum is what we notice. What we see, process, and react to is a by-product of both our unconscious associations with the “image,” what our unconscious is “interested” in, and what has been done to draw something to our attention. For example, without changing any of the products on offer within a store display but rearranging them to create visual impact, it is possible to increase sales dramatically, simply because people happen to notice what is there.
At the other end are the inner drives that provide the impetus to take action: these include the basic physical drives for food and warmth, the psychological desires for things like status and social contact, and the habits that create patterns of behavior that mean we don’t have to think any more than absolutely necessary.
In between is the complex interplay between contextual elements that influences us to perceive things in a particular way, and how we process what we encounter to decide whether it’s worth bothering with.
Increasingly, I find that it is from demonstrating and explaining these drivers of our behavior that organizations are able to see the importance of generating insights into their consumers’ unconscious minds.
Some people will accept the validity of the science available and will find that it helps to explain market research they have encountered themselves that was wide of the mark (remember, most market research failures don’t enter the public domain). However, not everyone has a mind that approaches matters in that way. I have found that taking people through an experience that reveals their own unconscious mind at work helps them see themselves in a different light; thereafter it is much easier to accept that other people do things for reasons they can’t adequately or accurately explain.
Recently, after attending one of the workshop-style sessions I use to illustrate the unconscious mind at work, a company director commented that I’d given him so much to think about that he wanted to go away and ponder it more. This request might have indicated that he hadn’t yet come to terms with the view of consumer psychology I’d been advocating, or that I’d gone into too much detail and overwhelmed him, but I knew that wasn’t the case. In the discussions we had after I’d demonstrated how consumers think, he frequently referenced the terms I’d introduced him to and astutely applied them to some of the commercial challenges we were discussing. In other words, while he wanted to consider what I’d talked about consciously, his unconscious mind had already implicitly accepted it.
I suspect that there are three reasons. The simplest answer is that many people are not aware of what behavioral insight is or how they could apply it.
A second potential barrier is that moving on from traditional market research techniques involves recognizing their flaws (if you still believe that they’re effective, why “risk” change?). This, in turn, requires at least a little willingness to recognize one’s own fallibility. The conflict (or cognitive dissonance) that arises from learning about the science that undermines market research and a history of having used it is most easily resolved by forgetting about the science!
The third barrier is a practical one and relates to the way in which most companies are structured. Market research techniques typically sit in a self-contained space away from an organization’s day-to-day activities. As you will have gathered by now, this stripping away of context is one of the reasons the answers obtained are frequently unreliable. Conversely, proper behavioral experiments may require different departments within a business to work together to generate insights; a simple communication test might require multiple letters to be designed and distributed and the responses delineated. A more complex test might have implications for marketing, operations, computer systems, human resources, and logistics functions. So the way organizations have evolved into separate functional areas can be a barrier to implementing the live testing that is so valuable to behavioral insight.
In many ways it’s a mistake to try to compare the costs of the two approaches, given that one will be showing what people actually do and the other will, frequently, be demonstrably unreliable. It is a little like suggesting that a homeopathic consultation is cheaper than a course of chemotherapy. That said, very often behavioral insights can be obtained at far less expense than is involved in market research surveys and focus groups.
It is somewhat ironic that while many people and organizations are still only just starting to consider the potential benefits of moving on from traditional market research and using a behavioral approach, others are asking whether it is too successful to be ethical.1 What critics of behavioral insight appear not to consider is that all communication is a matter of influence. Every message, however dull and factual, is presumably being delivered with the intention of people taking notice of it. Finding a means of engaging with people such that more of them act in accordance with your message doesn’t necessarily mean that you are coercing them into action that they wouldn’t otherwise take. Arguably, the mistake is to believe that there is a rational, objective reality on which a person’s actions will be based.