Write the tasks before the script, keep them goal-shaped rather than instruction-shaped, and recruit people who resemble real users of that flow. Five to eight participants per distinct group surfaces most severe issues. Ask people to work as they normally would, stay quiet, never lead, and record task success, hesitation points, and quotes. Then rank findings by severity and frequency.
Why interviewers ask this
Usability testing is the most common research skill on a product team, and it is easy to do badly in ways that produce confident nonsense. Interviewers listen for task design, because leading tasks invalidate everything after them, and for how you avoid coaching participants. They also want to hear how findings become prioritized work rather than a forty-slide deck nobody reads.
How to structure your answer
- Define the decision and write goal-based tasks first.
- Recruit for the behavior you are testing, not just demographics.
- Explain how you stay neutral and handle a stuck participant.
- Score severity and frequency, then convert findings into tickets.
- Invite the team to watch at least one session live.
Example answer
Tasks first, script second. A bad task is 'click the settings icon and change your notification preference', because you have just done the hard part for them. A good one is 'you are getting too many emails from this product, sort it out'. Then I recruit for behavior. If I am testing an admin flow, I need people who administer things, not whoever is available. Five or six per group gets me the severe problems; if I have two very different user types I run five of each rather than ten mixed. During the session I mostly keep quiet. When someone stalls, I wait, and if they ask me what to do I say what would you do if I were not here. That silence is uncomfortable and it is where the findings live. Afterwards I rate each issue on severity and how many people hit it, and anything severe and frequent goes straight into the backlog with a clip attached. Getting the PM and one engineer to watch a live session is worth more than any report I could write.
Walking into this interview soon? GhostPilot listens to your live call, spots the question the moment it is asked, and puts a structured answer on your screen in real time. Try it on your next mock, or grab a $29 Session Pass, no subscription, for the real thing.
See how it worksFollow-up questions to expect
- How do you handle a participant who cannot complete a task at all?
- How do you decide what counts as a severe issue?
- What do you do when the findings contradict what your stakeholder wants?
Related ux designer questions
Your interviewer will ask their own version of this. Paste your actual job description into the free Question Predictor and get the 20 questions that role is most likely to ask, with what each one is really probing.
Predict my questions