Don’t ask research questions from customers what to do instead
Creating an interview guide is simple: Open a new document and write down questions on what you would like to learn from your customers. Or that’s what many people think. They think they can just ask research questions from their customers.
Research questions tell you what you want to learn. But they don’t tell you how you should go about learning it. They’re not formulated in a way that helps uncover insightful, unbiased responses from customers.
Turning research questions into interview questions
Let’s imagine I work for an online grocery store that wants to better understand the drivers and obstacles to purchasing food online (because if we do, perhaps we can figure out a way to convert more people to using our service). Below you can see my team’s research questions as well as the interview questions I’ve created based on them.
Could you tell me a little bit about yourself? Potential prompts: family members, pets, work, leisure, etc.
Next, I’d be curious to learn a bit about your grocery shopping habits. If you think about the past seven days, how many times did you shop for groceries (including both bigger hauls and smaller trips to the corner store)?
Could you tell me a bit more about the last time you shopped for groceries?
Potential follow-up questions: When was this? Was there someone with you? What did you buy?
Would that be a typical trip to the grocery store for your family? If yes/no → What about it was typical/atypical?
Have you ever brought groceries online? If yes → Could you tell me a bit more about the last time you did that?
How was your experience of shopping groceries online?
Potential follow-ups: Did it feel different from real-life grocery store trips? If yes → How was it different?
I’m not asking the obvious: “Why do you buy / not buy groceries online?” Instead, I try to learn more about the interviewee’s life (the context in which she acts and makes decisions) and the way in which she has recently shopped for groceries (her past behavior).
I’m tiptoeing around the topic, because we humans are not very good at identifying the true reasons for our behavior.¹ We’re good at telling stories though: if you ask someone why, they’ll tell you why. But the response may have little to do with why they behaved the way they did; they may just be telling you what’s currently most accessible to their conscious mind.²
So if you want information that can actually help you solve the design problem, ask customers to tell you about their past behavior — not the beliefs and attitudes they hold about their past (or future) behavior. Fish for stories about the past, and if you get a good catch, keep dipping your rod into that same spot for a little longer.
And don’t shun away from specificity. The more vivid the story, the easier your job — especially, when it’s time to share the insights with your team.*
Alright, alright, but we still need to answer those research questions!
Yes, once you’ve collected a sufficient number of stories — with this technique, similar issues usually start popping up after about five to eight conversations—study them and see if you can detect any recurring patterns or themes. And then try to answer your original research questions: What are the drivers to shopping groceries online? What are the obstacles?
Answering research questions is the job of the researcher (and her team). So don’t go about asking research questions from customers.
Creating an interview guide isn’t as simple as it may first appear. But if you understand why research questions don’t equal interview questions and focus on uncovering specific stories about past behavior, it’ll turn into a stimulating brain teaser.
Reference:
- Nisbett, R. E., & Wilson, T. D. (1977). Telling More Than We Can Know: Verbal Reports on Mental Processes. Psychological Review, 84(3), 231.
- Wilson, T. D., Hodges, S. D., & Lafleur, S. J. (1995). Effects of Introspecting About Reasons: Inferring Attitudes From Accessible Thoughts. Journal of Personality and Social Psychology, 69(1), 16.
