Questions to Ask a Designer in an Interview
Questions for interviewing a designer, written for the person doing the hiring. They dig into portfolio ownership, decisions under constraint, work with engineers, handling feedback, and where the candidate knows they are weak.
The questions
Open any question for the note
Walk me through a project you are proud of. What was the brief, and what did you actually do?
Why ask it
The word "actually" is doing the work. Strong candidates separate their contribution from the team's without being prompted; weaker ones narrate the whole project as though they made every decision.
Who else was on that project, and what did they own?
Why ask it
This catches inflated portfolios politely. A designer who can name the researcher, the engineer, and the product manager and describe what each contributed has worked in a real team; one who cannot usually worked alone or is claiming too much.
What was the problem you were solving before you knew what the screen should look like?
Why ask it
Some designers start from a layout and work backwards to a justification. Listen for whether the problem is described in terms of what users were failing to do, and whether it is stated in a sentence that does not mention an interface.
Show me something that did not work out. What happened?
Why ask it
The most useful answer in any design interview. Look for a specific mechanism of failure and a lesson that changed later work. Candidates who blame stakeholders entirely, or claim nothing has ever failed, are telling you how the next failure will be handled.
How did you know whether it worked after it shipped?
Why ask it
Many designers hand off and never look back. Answers that mention a metric, a support ticket volume, a follow-up study, or even a single observed session put them in a small minority.
Tell me about a time you were handed a solution instead of a problem. What did you do?
Why ask it
This happens constantly and reveals how they operate without authority. The interesting middle ground is a designer who built the requested thing while quietly getting evidence for the alternative, rather than one who either refused or complied silently.
How do you decide a design is finished?
Why ask it
Perfectionism and carelessness both show up here. Look for a stopping rule tied to something outside their own taste: a review, a threshold of open questions, a ship date they helped set.
What is your process when there is no research available and no time to get any?
Why ask it
The honest answer involves proxies: support logs, sales calls, analytics, five minutes with anyone who talks to customers. A candidate who says they would insist on research has not worked somewhere that would not give it.
How do you hand work to engineers, and what do they usually come back to you about?
Why ask it
The second half is the tell. Designers who genuinely work with engineers can list the recurring gaps, empty states, error states, loading, long strings, and often say they have started specifying them up front.
Tell me about feedback you disagreed with. How did it end?
Why ask it
You are looking for a case where they changed their mind, and a case where they held a position with evidence. Someone who only has stories of being proved right is either lucky or not remembering accurately.
What do you do when a stakeholder starts redesigning your work in the review?
Why ask it
This is the most common failure mode of design reviews. Good answers usually involve reframing to the goal or offering a structured choice; answers that involve enduring it, or ignoring it and shipping anyway, both predict trouble.
How have you handled it when the accessible option and the more attractive option conflicted?
Why ask it
Anyone can state that accessibility matters. Ask for a specific instance, contrast, focus states, target size, and see whether they have ever actually spent a design decision on it.
How have you worked with a design system, and when did you deliberately break it?
Why ask it
Both halves matter. A designer who has never departed from a system may be a component assembler; one who departs constantly makes work nobody can maintain. Listen for how they decided and who they told.
How do you present work to people who do not speak design?
Why ask it
Watch how they explain it to you in the answer itself. Reliance on words like clean, modern, and intuitive is a signal that stakeholders will not understand their reasoning either.
Which part of the craft are you weakest at right now?
Why ask it
A candidate who names something real, motion, typography, writing, research method, and says what they do about it is easier to manage than one who offers a strength in disguise. The evasive answer is itself the answer.
Tell me about a deadline you were not going to make. What did you cut?
Why ask it
Scope judgement separates senior designers from thorough ones. Look for a decision about what mattered least to the user, made early enough to be useful, and communicated rather than absorbed.
What is the last thing you learned that changed how you work?
Why ask it
This distinguishes curiosity from consumption. Naming a specific change in practice is worth more than a list of newsletters, conferences, or tools they follow.
How do you want feedback on your own work?
Why ask it
Cheap to ask and immediately useful if you hire them. It also surfaces past experience: people who ask for written feedback in advance, or for problems rather than solutions, usually learned that from a review process that hurt.
What kind of team do you do your best work in?
Why ask it
Compare the answer honestly against the team you actually have. A designer who needs a research partner and close engineering collaboration will struggle as the only designer in a shipping-first company, however good the portfolio is.
What do you want to be better at in a year, and what would you need from us?
Why ask it
This tests whether they have thought about the role rather than the job title, and it tells you what retaining them will require. Answers with no ask usually mean they have not decided they want this job.
Running a designer interview
Practical guidance for the conversation itself
Structure the hour
Portfolio walkthrough, forty minutes
One project, in depth, chosen by them. Interrupt for decisions rather than waiting until the end: why this and not that, what you tried first, who disagreed. Depth on one project beats a tour of six.
Situational questions, fifteen minutes
The stakeholder, deadline, and handoff questions above. These are about how they behave under constraint, which is most of the job and none of the portfolio.
Their questions and needs, the rest
Leave real time. What they ask about, process, research access, engineering relationships, tells you what they have been missing and whether you can provide it.
How to press without interrogating
- When they describe a decision, ask what the alternative was and why it lost.
- When they say the team decided, ask what they personally argued for.
- When they mention research, ask how many people and what surprised them.
- When a screen looks finished, ask what the error and empty states do.
- When they cite a metric, ask what it was before and who measured it.
- If an answer stays abstract twice, ask for a single specific instance instead of rephrasing.
Common mistakes by interviewers
Judging the visuals and calling it a judgement of the designer
Polish is partly a function of how much time and freedom the last employer allowed. Ask what constraints the work was made under before you read the craft as the ceiling of their ability.
Asking for a design philosophy
It reliably produces rehearsed abstraction. The same information comes out of asking how they decided one specific thing, and that version cannot be prepared in advance.
Unpaid take-home exercises on real problems
Strong candidates decline these, so you filter for availability rather than skill. If you need to see them work, run a short paid exercise or a live session on a problem you have already solved.
Only interviewing designers
The engineers and product managers who will work with them daily notice different things. Give one of them a slot and a specific question to own.