Now available: Design for Impact by Erin Weigel

Indi Young on generative versus evaluative research


As part of our ongoing series of short interviews with Rosenfeld Media people, we turn to Indi Young, the author of our first book, Mental Models: Aligning Design Strategy with User Behavior. Since its publication in 2008, the book has become a perennial favorite; we just keep printing more!

RM: What’s the biggest mistake people make when it comes to mental models?

Indi Young: Someone said to me last year, “So I still don’t understand how to set up a scope to find out why someone would buy an iPhone over a Windows phone or an Android phone.” That would be evaluative research; mental models are a structure to contain generative research. The scope this person would explore generatively would be, “Tell me your thoughts around staying connected while on the go.” With a scope like this, you will hear the goals people have in mind—”Keep track of where the kids are,” “look up my client’s business address to find nearby parking,” “let my husband know I’ll be 30 minutes late getting home,” “change my flight reservation because this meeting is running long,” “listen to that new song my friend gave me after I get home, on my quality speakers,” “give that report to my co-worker after I finally remember during dinner at the pizza place,” etc., etc.

These are the things people are trying to get done, and they might use a mobile device to do it. Or they might use a phone book or the TV or a map or a piece of paper. It doesn’t matter. What matters is how each goal is supported. Does the Android phone support “give that report” as well as an iPhone or a Windows phone? You’d look at how each of these devices helps a person accomplish each of the goals. If there are stronger matches between certain devices and certain goals, then that indicates why a person might select the device with the stronger match. But just asking for the matches is a weak use of a mental model. Instead, use it to think up more specific, stronger ways to support a particular goal, for a specific subset of the audience.

Another mistake is the assumption that a lot of time and effort is required to create a mental model. Folks get scared off and never try one. True, if you go interview a bunch of people then yes, it will take six weeks or longer to get through all the steps. But you can also get a lot of insight using short essays people write about their thought processes regarding a particular scope. Or you can re-use existing research and your own knowledge of the customer perspective. These are a few ways to create, or at least sketch, a mental model within a week.