User Research interview questions
User Research questions a UX / UI designer can be asked, from first job to senior. Open one to see the strongest answer and why it works.
What must a research session have before you book people?
The strongest answer: A question you could be wrong about — not “validate our mock” as the only goal
Why it works: Exactly. A falsifiable question. Compliment scripts and promised 90% are theatre.
- Start with a question you could lose.
- Do not only validate a mock.
- Teammates ≠ users.
- No promised preference %.
The loudest voice in the room is not the person who will use it. What do you do?
The strongest answer: Bring a user moment into the room — a quote, a clip, a task — I do not decorate the loud voice’s taste
Why it works: Exactly. A moment in the room. Taste, ignore-both, and invented matching personas skip the user.
- Bring a user moment into the room.
- Do not decorate the loudest taste.
- Do not ignore both.
- Do not invent a matching persona.
A survey asks “How much easier is our new checkout?” What is wrong with it, and what do you send instead?
The strongest answer: It assumes the answer and leaves no way to say “harder” — I ask what they were trying to do and what happened, with a balanced scale if I need a number
Why it works: Right — a question that cannot return bad news is not measurement, it is decoration.
- A question must allow a “no”.
- Do not assume the verdict.
- Ask what happened, not how great.
- Balance every scale.
How do you start a user interview about checkout?
The strongest answer: “Tell me about the last time you tried to pay for something like this” — not “would you use our new stepper?”
Why it works: Exactly. Last time. Demos, love-scores, and grocery-NDAs collect politeness or silence.
- Ask the last time they did the job.
- Do not lead with your stepper.
- Love scores are not stories.
- Do not demo-first for compliments.
You have one week. How do you plan research?
The strongest answer: One question, a few of the right people, a synthesis you will use Monday — I do not book twelve methods to look thorough
Why it works: Exactly. One question, usable Monday. Buffets, leading surveys, and Slack-as-users skip the week.
- One question, right people, Monday artefact.
- Do not buffet methods.
- Do not lead the survey.
- Slack ≠ users.
Eight people told you the export flow is fine. Support tickets say otherwise.
What do you do next?
The strongest answer: Watch a few of them actually export something — people report the version of the day they remember, and the workaround they built is usually invisible to them
Why it works: Exactly — the say-do gap is why observation earns its place next to interviews.
- Watch the task, do not only ask.
- Book a session on their own file.
- Note the workaround they never mention.
- Compare it with what they said.
You have eight interview notes. What is synthesis?
The strongest answer: Patterns that answer the question, with quotes under them — a folder of recordings is not synthesis
Why it works: Exactly. Patterns for the question. Folders, dumps, and pies of eight are costumes.
- Patterns + quotes under the question.
- A folder ≠ synthesis.
- Do not dump transcripts as the artefact.
- Do not pie-chart n=8 as 100%.
Notes live in twelve Drives. What do you fund?
The strongest answer: A home tagged by question and decision, with consent — I do not fund an 80-page toolkit nobody opens
Why it works: Exactly. Findable by decision. Toolkits, bans, and memory are how you re-interview the same pain.
- Store by question and decision.
- Keep consent with the notes.
- Do not fund a shelf toolkit.
- Do not ban notes to stay light.
Your segment is a nurse on a shared ward tablet, gloved, working between patients. What changes in the screen you draw?
The strongest answer: Bigger targets, almost no typing, a task that survives an interruption, and nothing personal left on a shared screen — the constraints change the layout, not the poster
Why it works: Right — a persona earns its keep only when its constraints show up as decisions you can point at.
- Turn the person into constraints.
- List device, hands, interruption, privacy.
- Change layout, input and saved state.
- Point at each decision it caused.
A team says the job is “use our reporting tool.” Why is that not a job, and what does a real one show you?
The strongest answer: A job is the progress someone wants in a situation — “when the board meets, show what changed and why” — which reveals the real rivals: a spreadsheet, an analyst, or doing nothing
Why it works: Exactly — naming the situation and the wanted progress shows who you are really competing with.
- A job is progress in a situation.
- Name the trigger moment.
- Name the wanted outcome.
- List what does it today instead.
How do you make research more inclusive without a fake diversity %?
The strongest answer: Recruit for the jobs and constraints you have been missing, and pay people — I do not quote a diversity percentile I invented
Why it works: Exactly. Missing jobs, pay. Invented 40%, stock photos, and exposure-as-pay are costumes.
- Recruit the missing jobs and constraints.
- Pay people.
- No invented diversity %.
- Stock photos ≠ inclusion.
A slide says NPS will rise 12 points. You have no study. What do you do?
The strongest answer: Take it off — say what we will watch, I do not keep a costume 12
Why it works: Exactly. No study, no 12. Likes, decimals, and sourceless industry are worse costumes.
- Remove invented NPS lift.
- Say what you will watch.
- Leadership likes ≠ evidence.
- No sourceless industry NPS.
Designers are rewarded only for shipping screens. What breaks?
The strongest answer: They will skip the question — I reward a killed idea and a well-asked study, not only a launch file
Why it works: Exactly. Incentives write the craft. Ship-only, Dribbble-quotas, and screen-counts are how you get junk.
- Reward questions and honest kills.
- Not only launch files.
- Do not quota Dribbble.
- Do not run a screen-count board.
What does inclusive-by-default mean in a staff critique?
The strongest answer: We ask who is missing from the job and the test, and we do not treat a stock persona as the answer
Why it works: Exactly. Missing jobs and tests. Invented %, stock everyone, and date-skips are costumes.
- Ask who is missing from job and test.
- Stock personas ≠ the answer.
- No invented diversity %.
- Do not skip for a date.
What consent bar do you set for research?
The strongest answer: People know what we record, how long we keep it, and that they can stop — I do not hide a session in a “casual chat”
Why it works: Exactly. Know, keep, stop. Hidden recorders, forever-clips, and invented legal years fail people and counsel.
- Record, keep, stop — said out loud.
- No hidden recorders.
- No forever clips for the brand.
- Do not invent a legal year.