Interview guide for designers: a full loop with a portfolio review and a live design exercise

Designers arrive with a portfolio, which feels like better evidence than most candidates bring. It is also easy to misread. A portfolio shows finished work, often shaped by a team, a design system and a product manager, and it rarely shows how the designer handled ambiguity or feedback. A good designer interview loop uses the portfolio as a starting point and then watches the candidate frame a problem, explore options and take critique. This guide lays out a loop for a mid-level product designer: five stages, who owns which competency, core questions with what to listen for, a live design exercise, example scorecard weights and the debrief decision rule.

The recruiter's first call, built around one portfolio case study, is in UX designer phone screen questions. For visual, brand and production designers, see graphic designer screening questions and the adjustments at the end of this page.

The loop at a glance

The two design stages own different things: the portfolio review is about past reasoning, the exercise about present craft. Keep them apart so the same interviewer is not scoring the same thing twice. The logic is in the interview process template.

Stage 1: the screen (30 minutes)

• "Of the projects in your portfolio, which one is closest to what you would do here, and what was your part in it?" Listen for: a clear line between their work and the team's, and a project type close to yours (for example, a complex workflow tool rather than a marketing site).
• "Who did you work with day to day: how many product managers, engineers and researchers?" Listen for: the ratio they are used to. A designer who has always had a dedicated researcher will work differently on a team without one.

Stage 2: portfolio review (60 minutes)

Ask the candidate to present one or two projects for 25 minutes, then use the rest for questions. Tell them in advance that you want the process and the decisions, not only the final screens.

1. "What was the problem when it reached you, and how did your understanding of it change?" Listen for: a problem stated in terms of user or business outcome, and a moment where research, data or a conversation changed the direction.
2. "Show me an option you rejected. Why did it lose?" Listen for: a real alternative with a reason tied to the problem, not "the stakeholder preferred the other one".
3. "What shipped, and what happened after it shipped?" Listen for: an outcome they measured or observed, and honesty when the result was mixed or unknown.
4. "Where did you disagree with the product manager or engineer on this project?" Listen for: the trade-off itself, how it was settled, and whether they changed their mind for good reasons.

Stage 3: the live design exercise (75 minutes)

The exercise is live, collaborative and invented. Use a problem in your product's general domain but not on your roadmap, so nobody can mistake it for free work. Let the candidate use their own design tool or a whiteboard; fidelity is not the point.

The brief

• The problem (5 minutes): for example, "Clinic staff need to reschedule a patient's appointment while the patient is on the phone. Today it takes six screens." Give two facts about the users and one constraint (it must work on a tablet at the front desk).
• Framing (15 minutes): the candidate asks questions; the interviewer answers as the product manager would.
• Exploration (35 minutes): at least two different approaches, sketched roughly, then one taken further.
• Constraint change (10 minutes): "Engineering says the calendar component cannot change this quarter. What now?"
• Wrap-up (10 minutes): "What would you test first with real users, and how?"

What to listen for

• Framing: questions about who, how often and what goes wrong today, before drawing.
• Exploring options: genuinely different approaches, not two versions of the same layout.
• Interaction craft: fewer steps for the common case, clear states for errors and edge cases, and sensible defaults.
• Accessibility: whether they raise contrast, touch target size, keyboard use or screen readers without being prompted.
• Response to the constraint: a workable revision, not frustration or a request to change the constraint.

If you prefer a take-home, the rules for capping and scoring it fairly are in how to evaluate a take-home assignment. Do not run both.

Stage 4: cross-functional interview (45 minutes)

A product manager and an engineer split the time and score collaboration from their own side.

• Product manager: "Tell me about a time the deadline meant shipping something you thought was not good enough." Listen for: a scoped-down version they could stand behind, a plan for the follow-up, and no sulking.
• Engineer: "How do you hand off a design? Walk me through the last one." Listen for: states and edge cases specified, availability during build, and reviewing the built version before release.
• Engineer: "Tell me about a design you changed because of something an engineer told you." Listen for: curiosity about technical constraints and a better design as the result.

Stage 5: hiring manager interview (45 minutes)

1. "Tell me about the hardest critique you have received on your work. What did you do with it?" Listen for: what they changed, and whether they can separate the critique from the person.
2. "Give me feedback on this screen." Show a real but flawed screen from a public product. Listen for: feedback tied to the user's goal, priorities rather than a list, and a tone a colleague could hear.
3. "Tell me about a project that did not achieve what it was supposed to." Listen for: their own part in it and what they do differently now. Questions that pair well with this are in interview questions for creativity.

Scorecard competencies and weights

Example weights for a mid-level product designer. Set yours at intake and lock them before the first candidate.

Build and print the sheet in the scorecard builder; it checks that the weights add up to 100. Use the same 1–4 anchors as the rest of your loops: 1 no evidence, 2 vague or team-level, 3 specific personal decisions with a reason, 4 the same plus a changed practice.

The debrief decision rule

1. Scorecards first, with evidence, before anyone reads anyone else's. Attach photos of the exercise sketches.
2. Gate: interaction craft at 3 or above from the exercise. A product designer who cannot design a clear flow under light pressure will struggle on real work.
3. Floor: no 1 on collaboration. A strong designer who cannot work with engineering ships less, not more.
4. Weighted total: in this example, 2.8 or higher on the 1–4 scale is an offer.
5. Portfolio polish is not a tiebreaker. If two interviewers disagree, go back to what the candidate said and did in the exercise and review, not to how the portfolio looked.

Adjusting for other design roles

Common mistakes in designer loops

• Scoring taste. "I liked their style" is not evidence. Tie every score to a decision the candidate explained.
• Unpaid real work. Asking candidates to redesign your actual product invites suspicion and gives you nothing you can fairly compare.
• Only designers on the panel. Product and engineering partners notice collaboration problems that designers miss.
• Penalizing rough sketches. The exercise tests thinking; fidelity in 75 minutes mostly measures tool speed.

After the offer, the 30-60-90 day plan for UX designers covers the first three months.