Two minutes before my first-ever user interview was supposed to start, my wifi died. Cue panic: me, furiously running around my house, restarting things at random and shouting at everyone.
I’ve been working on a project in which we’re building out websites for small businesses within certain niches. We started with restaurants, and I wanted to know as much as we possibly could about restaurant owners—what they struggle with, what they want and don’t want in a website, and how we can help to solve their problems.
So we embarked on a research project: we did a whole series of exercises to build out theoretical personas, and then we tried to validate those assumptions. First we tried surveys, but they didn’t yield much useful information. Interviews, however, gave us way more useful (albeit still a bit anecdotal) data to work with.
I’m still working through all the data, but here’s what I learnt about the process of actually running the interview itself…
1. Write a script, but recognise that you won’t follow it.
You’re going to go off-script. That’s okay. That just means you’re listening and responding to the other person. Your script is just there as a framework, but once people start talking, a lot of new questions will come up.
People really love to talk, especially about a business that they’ve put their blood, sweat, and tears (and often rather a lot of money too) into. This really shouldn’t have come as a surprise to me—I ran a business for eight years and I would have prattled on endlessly had someone sat me down to talk about it—but I admit I thought it might be a bit harder than it was to get people feeling comfortable and sharing with a total stranger.
Nope! Everyone was super chatty and elaborated a great deal. I told people the interview would take 30-60 minutes but some of them actually ended up running longer than an hour. I think my longest interview ran a solid two hours straight.
2. It’s difficult to avoid leading questions.
At least once an interview, I’d catch myself accidentally asking a leading follow-up question. I think to some degree this is natural—the answer has already been suggested by the person I’m interviewing, and I’m trying to get clarification—but increased vigilance here would be helpful.
Sometimes I worry if I rephrase the question to be not-leading, then I’ll end up sounding like I haven’t been listening fully, or I wouldn’t be asking such stupid questions. Paying more attention to the questions I’m asking and how I’m phrasing them is something I think I’ll learn in time.
3. Reserve a bigger block of time than you think you need.
In the future, I’ll block four hours for each interview in my calendar. This allows for getting to the interview and/or setting up, running the interview, and then debriefing afterward, plus extra time spent writing up notes.
All of this took time I didn’t fully plan for, being busy with other things at the same time, and it would have been a bit smoother had I been able to block off sufficient chunks of time to devote solely to the interviews.
4. In-person interviews are better, and also worse.
I offered our participants the option of doing the interview however they were most comfortable: phone, video, email, text, or in-person (if they were local to me). Both of my local participants opted for the in-person option. Of the other three participants, only one chose video—the other two phoned into the Zoom chat I had set up.
I loved that we gave a range of options depending on what people were comfortable with, but I think the phone option may have been the worst of all possible worlds. In one of them, the unavoidable delay that happens with video chats was so pronounced that I ended up talking over my interviewee all the time and coming off super rude.
The in-person chats were great since you could form more of a personal connection—and it gave me the opportunity to see my interviewee in their restaurant setting—but were definitely more time-consuming, both in terms of travel and note-taking.
5. Have a note-taker or two.
I acted as my own note-taker for the in-person interviews I conducted, and it was definitely a bit more stressful (and time-consuming!) In this case, it actually helped quite a bit that the restaurants were busy, and I was able to scribble out notes when they were busy serving customers.
I was lucky to have Ola Bodera and Michelle Langston volunteer as note-takers for my other interviews, which took a huge burden of responsibility off me and allowed me to fully focus on the question-asking. (Thank you both so much!)
5. Attach a photo.
There’s so much data that’s gleaned from an hour-long interview, it’s a bit overwhelming to figure out what to do with it.
I’m starting with the individual stories—combing through the recordings and the research notes, trying to compose a story of every person I interviewed. (The stories are already there, I just need to pull them out.) Chatting with our business owners made me more invested in their business’ success, and I think sharing those stories will help make others feel invested as well.
There’s this study that was done in New York where they found radiologists studying CT scans gave far more thorough reports when a photo of the patient was attached to the scan, which is pretty wild if you think about it. They didn’t spend any more time examining the scan, they were just more thorough—they cared more because they attached a face to it. That’s a pretty powerful effect that we can harness ourselves, so I’m attaching photos to every story. Luckily, as part of the pilot project we also sent around a photographer to each of our restaurants, so we have photos of our owners to work with.
6. Finally, make sure your wifi is working.
It’s going to be a really lousy interview otherwise.
Automattic Design 