Creating Realism for More Fruitful User Interviews

As UX professionals, we are accustomed to engineering experiences for others, and we tend to bring this same strategy into user interview situations. There are important reasons for doing so; from a methodological perspective, it allows us to control as many variables as possible so as to reliably isolate and examine the variable in question. We also do it to avoid obscuring or changing a participant’s real behavior. Some simple and common practices that we follow at User Centric include paying attention to consultant dress (suits are off-putting), age (pairing younger-looking moderators with teenage subjects) and gender (male moderators for men’s health studies). In the lab, we often take the pursuit of realism much further, from hiding our recording equipment from view and offering cups of coffee to building entire sets to resemble hospital rooms, complete with accurate lighting, ambient sounds, and furniture. 

There are many creative examples of built environments for user testing (check out the research process that resulted in the Air New Zealand Skycouch for an example that will blow your mind), but they still utilized a lab setting, where the researcher had a lot of control over the context. 

What happens, then, when researchers cede some of that control by going into the field?

As researchers, we sometimes mistake things like home or office visits for more natural environments in which to do research, but the very presence of the researcher brings a similar power dynamic into those settings. Maybe you’ve been there. A stranger/quasi-guest sitting on the couch in someone’s house at 10am on a Tuesday morning. The camera is running. They’ve signed their consent form. The place is a mess (you’ve asked them to leave everything in situ, for context) but you aren’t sure what last night’s dishes on the coffee table have to do with streaming media. “Participant eats in front of the TV” you jot down on your notepad. Now it’s time to Ask The Questions. They don’t have much to say about their reasons for purchasing a Roku, and the interview never really catches a spark after that. But you’re in the field, right? Isn’t that supposed to mean heady insights? The fact is, sometimes simply going into a participant’s home or office space for an hour is not sufficient for making a connection or otherwise achieving a genuine state of context, both of which are critical to getting good information.

Ethnography, practiced in its purest form, is facilitated by the unabashed optimism that throwing oneself as far outside the zone of comfort and predictability as possible will ultimately yield the most fruitful insights—exactly because they could never have been anticipated or planned for. It is, in this way, the formative bench research of social science. The rub is that the products of ethnographic research are wholly dependent upon time. Like any good soup, the longer the ingredients are left in the pot, the more you can taste the flavors mingling in new and delightful ways. (This is why it takes PhD candidates in anthropology an average of 10 years to complete their dissertations. “Being There” might be the most popular title in the anthropological literature.)

Of course, we all know that business does not travel at the speed of nomadic tribes. So how do we bring the power of ethnography to bear more fruitfully on UX projects?

Negotiate for a little more time

This doesn’t always mean spend the whole day with your subject. Consider repeat visits. Get to know them a little.  People will often have very different moods and opinions on different days. Give them your number and tell them to call if they think of something. Self-reflection doesn’t always arise on schedule.

Give up some control

Don’t pre-script your questions. Don’t ask everyone the same questions. Let each interview help you identify other lines of inquiry for the next interview. Ask them to recommend other people you should talk to, and let this process guide your recruitment.

Think like a reporter

Go to the environments where you will observe and encounter the people you are interested in. User Centric did a kiosk study with an arcade bar & restaurant a few years ago. The guests used the kiosks to buy or load their arcade cards with points.  There was a lot of complicated math to decide what the best “value” was for the money, and the decision was usually a group decision of several friends or the whole family gathered around the kiosk.  We recruited groups of participants straight off the arcade floor and they often were noticeably buzzed or drunk, which made for a tough job moderating. But the reality of the situation helped us understand the context of how the system was actually used.

Let the formality go

Consider skipping the signed consent forms, and just explain honestly why you are there and what you want. Verbal consent is OK.

Use the right people

Encouraging abundant, honest, and real feedback in user interviews sometimes requires enlisting the help of someone who can do it better than you can. Anthropologists call them key informants. In South Africa, where research participants can hail from dozens of different tribes, UXalliance member Mantaray-IT uses cultural experts as moderators, who interpret not only dialect, but colloquialisms and non-verbal communication. Other UXalliance members have tried inviting participants to talk to their hairdresser, stopping into a bar and chatting people up over beers, and even bringing actors in to set the scene and act out relevant characters for the participants to react to.

In formative field studies, the more power the researcher is willing to give up in controlling their subjects and the context, the more likely they are to bump up against unexpected but genuine insight. Ideally, more contact time is required to create the surface area over which something can actually transpire. But even in the fast paced-world of UX, there are creative ways to create realism for more fruitful user interviews.

Another good read: The Underutilization of Ethnography in User Experience Research