Posted by Gina Sverdlov on May 30, 2012
Recently, I attended the MSI Workshop on behavioral economics at the Harvard Innovation Lab. The presenters included an innovative crew: a number of academics from Harvard Business School and Dan Ariely , author of Predictably Irrational. They gave many examples of how consumers aren’t always rational and don’t always know why they do what they do. This is troubling for market researchers, since it’s our job to understand what drives consumers so that companies can effectively optimize what works and what doesn’t. Let’s look at a couple of examples of how principles of behavioral economics can wreak havoc on market researchers:
- The path of least resistance (AKA the default). Ever wonder why you are enrolled in your company’s 401(k)? Why you take generic over brand-name prescriptions? Why you’re an organ donor? Is it because of your financial sense or your values — or could it be because it was simply the default (and hence the simplest) option? In fact, multiple experiments in behavioral economics suggest just that: People tend to go with the default option but don’t realize they do.
- Phrasing counts. John Gourville of Harvard Business School shared some results of a survey that was fielded to capture peoples’ likelihood of making a donation: Respondents were much more likely to say they’d make a donation of a dollar a day for a year than a lump sum of $350. People trivialize a dollar a day as an unsubstantial amount — but $350 sounds like a lot of money! This is what behavioral economists call framing.
So how can market researchers control these biases? Well, you can’t prevent them altogether, but there are a few measures that you can take to limit them.
- Frame the question in a way that puts the respondent in the right frame of mind. It always helps to describe the situation to the respondent. For example, rather than asking, “What are your favorite clothing brands?” rephrase it to something like “You’re at the mall and you only have one hour. Which stores do you go to?”
- Derive how consumers make choices instead of asking directly. For example, in a quantitative survey, rather than asking respondents what is most important to their purchase decision, use quantitative analytics to correlate product-brand associations with where they shop most. In a focus group, rather than asking focus group members what they like about their cell phone, ask them to imagine a day without it and describe what they would miss.
- Understand that people will answer anything. Designing a survey with behavioral economics in mind results in much better questions. For example, with a survey question that lists five brands and asks respondents to select the two they are most likely to buy, respondents will feel obligated to select two brands from the list — even if they have no intention of shopping from any of the five brands.
Because consumers don’t always do what they say — or say what they do — there will always be biases in market research. But making improvements in the way we conduct research can be as simple as having an understanding of some of these limitations.