Interview Techniques for Users

I had a great talk the other night with a classmate of mine from graduate school, who focused on usability and now works on a web application development team as their user experience designer. He’s Tim Keirnan, and I asked him to explain some of his interview techniques that he uses for his Design Critique podcast. I also got great snippets about his user interviews.

I just marveled at how dedicated he is at getting user information, no matter what the situation. Just acquired a company in an eastern European country? No worries, set up a remote meeting where you can view the user’s desktop. Think your users don’t know what they need in their daily workflow? Ask the right questions with the right context and you’re halfway there at least.

  • Don’t ask people to project into the future. For example, asking “What will you do with the Intranet tomorrow?” sound ridiculous if you phrase it that way. Instead, ask about today, this week, this month.
  • Don’t ask them anything without having artifacts in front of them to spur discussion, even if you have to use Webex because you can’t travel to their office.
  • Do ask users about their daily work.
  • Do ask about their last mistake and what they think they could have done to prevent it.

These are just a few notes I jotted down after talking to him. Since most “interview techniques” posts are about job interviews, I wanted to find more user interview techniques. There are eight articles about conducting usability interviews collected on the EServer TC Library. One essay titled Nondirected Interviews: How to Get More Out of Your Research Questions on Adaptive Path says to concentrate on immediate experiences, which really hit home with me. At SXSW Interactive, much ado was made about context. And context in interviews – both time and place – can make or break the value of the user research.

I ponder this idea of context as I search for user data on the social web. Perhaps the loudest users are on social media but not a good representative of true users, depending on the product. I’m thinking of the information you can get about job titles, job descriptions, tools used, and so on from a site like LinkedIn or I described finding your user’s vocabulary previously in a blog post last year. What do you think? Should you emphasize and use the online personas you can build from social media sites, or call that publicly-available information suspect and marginal?


  • April 7, 2010 - 9:10 am | Permalink

    I guess, as you suggest, the answer to your question depends on the product and its user community. If you know that your users are tech savvy and likely to be plugged in to social media, then you probably can count on the social web as a reliable source of information. You could even use, say, a LinkedIn group to pose questions like “How do you do this task?” or “What’s your most common mistake?”

    I love the point about not asking users to project into the future. They might think they know but they don’t. If I’m designing a user interface I need to understand what my users are doing today, so that I can design my workflow and my UI in such a way that they’ll look like something the users already know.

  • Techquestioner
    April 19, 2010 - 3:46 pm | Permalink

    I think that if you are plugged into a variety of social media sites and get to know groups/categories of users from reading their posts and interactions regularly, it can help you create more descriptive personas. For instance, members of a Red Hat Club site may discuss their upcoming events, but they may also discuss things like getting ahead in their jobs, or going back to work after their kids have gone off to college. I get a lot of posts from my daughter’s college friends, who all came to our house for a costume project last summer. Many of them are going through the pre-graduation interview process, and posting about it. I could use that knowledge to create a persona for a new graduate job seeker. I think you have to start with the resources you already have, especially if you don’t get any resources specifically for user research.

  • April 26, 2010 - 1:08 pm | Permalink

    Here’s another great question to ask when doing user testing.

    Users frequently will suggest better ways to do something, “the search engine needs to let you sort by date”, “the font needs to be blue,” etc.

    However, users almost always give non-actionable suggestions because they lack the insight about how such a feature could be implemented or what the existing constraints are. Often, user suggestions are rejected by developers as being unrealistic& impractical & preposterous for this very reason.

    If a user gives a suggestion, you should immediately follow up with: “What problem is your solution supposed to resolve?” I can promise you: whatever comes from a user’s mouth at that point will be pure GOLD.

  • Leave a Reply