We met Diane Loviglio, a user experience researcher at Mozilla, at a recent SF Bay Area meetup, and she shared some insights on the traditional Recruit-Observe-Interview-Navigate user testing process.
1. Recruit your target audience. Diane suggests not talking to just anyone or everyone. Instead, recruit people based on “behaviors” or “personas” versus standard demographics. For example, if you target “homeowners” versus “married couples in their 30s”, you’re likely to gain more insights from the former group than the latter if what you wanted to learn was related to finding a home loan. Young married couples may or may not have gone through the mortgage process yet.
2. Observe users in their own environment versus a lab. When users are in their own environment, they use their own devices, data, keyboard short-cuts, bookmarks, browsers, etc. The researcher gets to observe real-world scenarios of problems that the user would likely encounter. If the lab was setup to support only one type of browser, for example, results from user testing may overlook some obvious issues.
3. Interview users with question words like “who, what, when, where and why”. It sounds obvious to ask questions with question words, but Diane has noticed many “yes/no” questions being asked during user testing interviews. For example, you’ll obtain more information by asking “how did you figure out how to get here” versus asking “did you use Google maps”. Diane mentioned how Walmart learned this the hard way. Walmart spent millions of dollars to renovate their stores with larger aisles, thinking that’s what customers wanted. Walmart had asked “would you like bigger aisles” instead of “what would you change about your shopping experience”.
4. The user is in control. What Diane means by this is to let the user talk you through your product (also called ‘think-aloud’). A user test is not about you giving the user a product demo or helping them find or learn how to use a feature. Give the user a task to perform and then let them complete the task, even if they stumble through the task, as those are the opportunities for learning and gathering insight about usability. Diane suggests having two researchers, one to interview and one to take notes. If possible recording the session lets you the researcher direct your full attention to observing and listening to the user.
We think Diane’s tips are spot on. Thanks, Diane! We also think that remote usability testing with TryMyUI help you the researcher to accomplish Tips 1, 2 and 4 quite easily with minimal effort. For writing better user tests per Diane’s Tip 3, check out our posts called Five Secrets to Writing a Great Remote Usability Test and Tips for Writing a Usability Test.