A seven-step framework
- Clarify the goal and constraints: whose goal, what success means, any limits.
- Choose a user segment and say why.
- List that segment's pain points and needs.
- Prioritise one or two pain points by impact and frequency.
- Brainstorm three or more solutions, from simple to ambitious.
- Choose one, with trade-offs, and sketch the core experience.
- Define success metrics: a primary metric and guardrails.
The three question types
| Type | Example | Watch out for |
|---|---|---|
| Design | Design a product to help elderly people stay connected with family | Jumping to features before choosing a user |
| Improve | How would you improve Google Maps? | Listing ideas without a goal or prioritisation |
| Favourite product | What's your favourite product and why? | Describing features instead of the user problem it solves |
Choosing a segment
The most common mistake is designing for everyone. Pick one segment and justify it: large, underserved, aligned with the company's strategy, or reachable. For “improve Spotify”, you might choose people who use it while running because the pain points — hands-free control, tempo matching — are specific and frequent. Naming two or three other segments you considered, and why you set them aside, shows judgement.
Worked example: improve a food delivery app
Goal: increase repeat orders. Segment: office workers ordering lunch, because they order frequently and churn when lunch is late. Pain points: unpredictable arrival times, decision fatigue, and group orders with colleagues being awkward. Prioritise arrival predictability (high impact, very frequent). Solutions: more accurate arrival estimates with a confidence range; a “deliver by 12:30” scheduled order that only shows restaurants that can make it; and office drop points with batched delivery. Choose “deliver by”, because it turns a vague promise into a commitment users can plan around. Metrics: repeat order rate among lunch users as primary; on-time rate and restaurant cancellation rate as guardrails.
Answering “what's your favourite product?”
This question looks casual and isn't. Pick a product you genuinely use, ideally one the interviewer may not know well so your explanation carries weight. Describe the user and the problem it solves before any features, explain why its design choices work for that user, name one thing you would improve and how you would measure it, and connect it to something about the company you're interviewing with. Avoid the obvious picks unless you have an unusual angle; a thoughtful take on a niche tool beats a generic one on a famous app. Keep it to five minutes, and be ready for “how would you grow it?” as the follow-up.
Metrics that make sense
- One primary metric tied to the goal — repeat rate, activation, retention, time to value.
- Guardrails for what could get worse — cancellations, support tickets, latency, revenue per order.
- Leading indicators you'd see within weeks, not quarters.
- How you'd test: an A/B test, a limited rollout, or a fake-door test to gauge demand.
How interviewers score it
| Dimension | Strong signal |
|---|---|
| Structure | A clear, signposted path the interviewer can follow |
| User empathy | Specific, believable pain points |
| Prioritisation | Explicit criteria for choosing what to solve |
| Creativity | At least one non-obvious solution |
| Judgement | Trade-offs acknowledged; scope kept realistic |
| Metrics | Success defined with guardrails |
Common mistakes
Skipping the goal, designing for every user, listing a dozen features without prioritising, ignoring the company's business, and finishing without metrics. Another is defending your first idea after the interviewer challenges it; adjusting thoughtfully scores better than holding ground. See product manager interview questions for the wider loop.
Practising
Practise one design, one improvement and one favourite-product question a day for two weeks, out loud and timed at 20–25 minutes. A free AI interview assistant such as CoPilot Interview can run these as mock interviews and score structure and prioritisation. For AI-focused PM roles, see AI product manager interview questions.
Disclosure: CoPilot Interview is our product, so weigh what we say about it accordingly. Everything else on this page applies whichever tool you use, or none.
Frequently asked questions
What is a product sense interview?
A product management interview that tests whether you can identify valuable user problems and design sensible solutions, through questions such as designing a product, improving one, or discussing a favourite.
How do I structure a product sense answer?
Clarify the goal, choose a user segment, list and prioritise their pain points, brainstorm solutions, pick one with trade-offs, and define a primary success metric with guardrails.
How long should a product sense answer take?
Usually 20 to 30 minutes, with the interviewer interjecting. Keep each step brief enough to reach solutions and metrics.
What is the most common product sense mistake?
Jumping to features before defining the goal and choosing a specific user segment, which makes prioritisation impossible.
Practise on real interview questions
Run 145 real interview questions across 10 tracks, answer out loud and see a structured answer to compare with yours. Free, no signup.
Try a practice question