Consumers Aren’t Psychic, So Why Do We Ask Them To Be?

You know how you’ve told yourself that you were totally going to repaint your dining room this summer? Or train for a marathon – or at least a 5k? Or how you were definitely going to sign up for that cooking class? And then you didn’t. It’s cool, you don't have to be embarrassed. Everyone does it. Because people are absolutely terrible at predicting their own behavior.

But we still see a lot of research where the key takeaway is entirely rooted in responses to survey questions asking people what they might do at some point in the future. If your insights are tied to speculative questions, your research might be bad.

These types of questions can make for a great marketing hook but often offer little to no actionable insights. It is entirely realistic to craft a survey where you get 83% of respondents saying they would consider purchasing your pineapple and spinach flavored toothpaste. This data point later turns into a pitch to investors: “there is significant potential for growth in the fruited-vegetable flavor profile in the oral care market.”  But in reality, consumers aren’t going to buy that product, not now or in the future. They’re going to buy mint, and your investors are going to lose money.

But, as the saying goes, you don’t have to take my word for it. There’s a solid body of research that supports the idea that people are really bad at predicting future behavior for a variety of reasons. For example, people have a bias towards optimism – especially toward positive or rewarding activities (like running that 5k) – and overestimate their likelihood of engaging in behaviors that will lead to those outcomes. 

We also have a tendency to base predictions of future behavior on present intentions, so if you’ve been thinking about cooking at home more, eating healthier, or cutting back on the take-out, then you might report that you would consider signing up for that cooking class. But you might also fail to consider the time commitment, the cost, or whether you really wanted to learn to cook after all.

And we get worse as the speculative decision becomes more abstract – Would you consider buying this hypothetical product?  Would you pay more for this yet to be implemented feature? Would you ever move to this distant location? Again, we tend toward optimism – I could use that product! That feature sounds great! We could enjoy living there!

You can generate some interesting findings from these types of questions, but you are unlikely to generate useful insights about action or behavior.

So why do we ask these questions to begin with? Because they are really quite satisfying, if unhelpful. They are quite likely to tell us what we want to hear – people do want your product, there is a market for it, you are a disruptor! The problem is, they won’t deliver actionable insights. They will, however, deliver bad research.