Balkrishna Attal

Home / Blog / Interviews

InterviewsSeptember 28, 2026·8 min read

Product Sense Interview Questions: A Framework, a Worked Example and 15 to Practise

The round everyone fears is a learnable skill. A five-step framework, a complete worked answer, and fifteen prompts to practise.

— views
Quick answer

Product sense interviews test whether you can turn a vague prompt into a real problem for a real user and make a defensible choice. Use five steps: clarify, identify users and problems, generate three different solutions, prioritise and pick, and state a success metric. Interviewers score empathy, focus, creativity and judgment, not brilliance.

Product sense is the round people fear most, because there is no formula and no right answer. That fear is usually misplaced. Interviewers are not looking for a brilliant idea. They are checking whether you can take a vague prompt, find a real problem for a real person, and make a defensible choice. That is a learnable skill, and this page shows you how.

What a product sense round actually tests

Behind every product sense question ("design a feature for X", "improve Y") the interviewer is scoring four things quietly:

User empathyDo you know who this is for?Problem focusDid you pick one real problem?CreativityWere your solutions distinct?JudgmentDid you choose and justify?
Notice that 'a clever idea' is only one of four boxes.

A candidate with a merely good idea but no clear user usually scores lower than one with an ordinary idea for a sharply defined user. Keep that in mind when you feel pressure to be original.

A framework you can hold in your head

1ClarifyGoal, scope, limits2Users &problemsSegments, pain points3Solutions3 distinct ideas4Prioritise& pickImpact vs effort5SuccessmetricHow you would know
Say the steps out loud so the interviewer can follow you.

1. Clarify

Ask two or three questions before you say anything else. What is the goal (growth, retention, revenue)? Which platform? Any constraints? This is not stalling. It is the first thing a strong candidate does, and skipping it is the most common way to lose the round.

2. Users and problems

Split the users into two to four meaningful groups, not demographics for their own sake. Then list the biggest pain point for each. Choose one group and say why: they are the largest, the most underserved, or the most valuable.

3. Solutions

Give three genuinely different ideas for the chosen problem, not three versions of one idea. Different means different mechanism: a product change, a content change, a policy or pricing change.

4. Prioritise and pick

Compare the ideas on impact and effort and pick one. Then name the trade-off out loud: what this choice gives up and what would make you change your mind.

Big bets(plan carefully)Do first(high impact, low effort)Skip(low impact, high effort)Quick tweaks(do if free)Effort →Impact →
The top-right box is where you want your chosen idea to land.

5. Success metric

Finish by saying how you would know it worked: one primary metric and one guardrail. Interviewers love this step because most candidates forget it entirely.

A full worked example

Prompt: "Improve a navigation app for two-wheeler riders in Indian cities." (An illustrative example, not a claim about any real product.)

Clarify: "Is the goal more daily active riders, or fewer riders switching to another app? I will assume retention among existing riders, on mobile, with no hardware changes."

Users and problems: Three groups: daily commuters, delivery riders, and weekend leisure riders. Commuters repeat the same route and mostly care about surprises (sudden closures, waterlogging). Delivery riders care about time per drop. Leisure riders care about scenic or easy routes. I will choose daily commuters: highest frequency and the most likely to switch apps after one bad trip.

Solutions: (a) A pre-departure alert that warns about disruptions on the user's usual route. (b) Riding-specific routing that avoids roads with no two-wheeler access or poor surfaces. (c) A "leave now or wait" suggestion based on expected congestion.

Prioritise: Alerts are high impact for commuters and cheap because the route is already known from past trips. Riding-specific routing is high impact but needs new map data, which is expensive. The suggestion feature is nice but affects fewer decisions. I recommend (a).

Trade-off and metric: "This helps repeat commuters and does nothing for delivery riders, which is a deliberate choice. I would measure the share of alerted trips where the rider took the suggested alternative and completed the trip, with a guardrail on notification opt-outs so we do not annoy people."

How to segment users without wasting time

Most candidates segment by age or gender and stop. That rarely produces different problems. A better rule: segment by what they are trying to do, how often, and what is at stake. Two people of the same age using the same app can have completely different jobs. A student ordering food during exam week and a parent ordering for a birthday are the same demographic and different problems.

Daily commutershigh frequencyDelivery ridershigh stakesWeekend riderslow frequency
Example segments for the navigation prompt. Frequency and stakes usually decide which segment to pick.

After segmenting, say why you are choosing one. "Because they use it every day and are most likely to leave after one bad experience" is a complete reason. "Because they seemed interesting" is not.

Getting to three genuinely different solutions

Under pressure, people produce three versions of the same idea. Force variety by changing the lever each time: one idea changes the product itself, one changes the information or content the user sees, and one changes an incentive, price or policy. For a problem like "first-time investors quit after one confusing screen", the levers might be a redesigned screen (product), a plain-language explainer shown at the moment of confusion (content), and a small first-investment bonus (incentive). Three clearly different bets are far easier to compare and defend than three synonyms.

What separates a weak answer from a strong one

DimensionWeakStrong
ClarifyingStarts answering immediatelyConfirms goal, platform and constraints first
Users"Everyone" or age bracketsSegments by job and frequency, then picks one with a reason
ProblemA feature idea disguised as a problemA specific pain, backed by an example or evidence
SolutionsThree variants of one ideaThree different levers, compared honestly
DecisionHedges across all optionsChooses one and names what it costs
MetricForgotten, or "engagement"One primary metric and one guardrail

The follow-up questions to expect

Good interviewers push after your answer, and how you handle the push matters as much as the answer. Expect: "Why that user and not the others?", "What would you build if you only had two weeks?", "What is the riskiest assumption in your idea?", and "How would you test it before building it?" A strong reply to the last one is small and concrete: a prototype, a fake-door test, or five user conversations. It should never be "build it and see".

A two-week practice plan

1Days 1-3LearnStudy the framework2Days 4-8DrillOne prompt daily, out loud3Days 9-12MockTwo live mocks4Days 13-14RefineFix your top habit
Two weeks of deliberate practice beats two months of passive reading.

Days 1 to 3, read the framework and one worked example. Days 4 to 8, answer one prompt a day out loud in ten minutes, and record it once. Days 9 to 12, do two live mocks with someone who will interrupt you. Days 13 and 14, find your single most repeated bad habit (usually skipping the clarify step) and drill only that.

Fifteen questions to practise with

Mistakes that quietly lose the round

Solution before usermost commonToo many ideas, no pickvery commonNo success metricvery commonIgnoring the trade-offcommon
Illustrative ranking based on what I see in mock interviews, not a survey.

Proposing a solution in the first thirty seconds. Listing five ideas and never choosing. Speaking for ten minutes without a single check-in. Finishing without a metric. And treating the interviewer as an audience instead of a collaborator: ask "does this direction make sense?" halfway through. It is a conversation.

Want to see how this connects to other rounds? Read the 50 question bank next, then the skills guide for the underlying habits.

Frequently asked questions

What is a product sense interview?

A round where you take an open-ended prompt, such as designing or improving a product, and show how you identify users, choose a problem, generate solutions and make a justified decision.

How long should a product sense answer be?

Roughly 15 to 25 minutes in a live round, including clarifying questions at the start and check-ins with the interviewer along the way.

How do I get better at product sense?

Practise on products you use daily: name the user, a real problem with evidence, three different solutions, your pick and your metric. Do it out loud with someone who will push back.

Is there one right answer in a product sense interview?

No. Interviewers evaluate the quality of your reasoning and the clarity of your choice, not whether they would have picked the same idea.

Want this sharpened for your specific situation?

I take on a small number of career calls and resume reviews each month.

See how we could work together
BA

Balkrishna Attal

A product leader with a background in civil engineering. Writes about product management careers, interviews, and the frameworks that hold up in real work.