I was reading an interesting book on decision making (“How we Decide” by Jonah Lehrer) recently and was struck by a fascinating experiment that he recounts.

The experiment was conducted by Timothy Wilson and has real implications for how we conduct research. Essentially, he took consumer report ratings of strawberry jam and used them to rank a set of different jams. Then he gave the jams to a group of undergrads and asked them “which jam is best?” Asked in this direct way, their answers were pretty close to the established ranking.

However, when he asked them to explain why they liked one jam better than another as they made their choice, then the correlation completely disappeared! This is because as the students tried to explain their choice they had to rationalise it, and this process distorted their decision making. They began to concentrate on features of the product that helped them explain their choice (e.g. texture) that they might not ordinarily care about or even notice.

In consumer research, we often ask people to explain why they do things and we know that there is a huge amount of post rationalisation in the way they answer. What is key from this though, is that they might actually change their mind as they form their answer, rather than merely try to retrofit a rationale.

When we write our discussion guides and arrange our research we should make sure that we leave enough time between asking for a preference and asking for a reason for that preference. It’s important to make sure that we get a good clean read of their true opinions. By providing time for both types of questions, maybe we can have jam today after all!