Back to Resources

Resources · Product Management

PM Interview Prep

PM interviews test 6 distinct skills — and most candidates prepare only for product design. Here's the complete framework for every type of question you'll face at Google, Amazon, Flipkart and startups.

The 6 PM Interview Skills — With Frameworks

🎨

Product Design

Design a product for [user group]. The most common PM interview question — tests user empathy, problem structuring and solution creativity.

Framework

Clarify → Segment Users → Prioritise Pain Point → Generate Solutions → Prioritise → Define MVP → Success Metrics

Never start with solutions. Spend 30% of your time understanding the user before proposing anything. Interviewers notice when you rush past the problem.

📈

Metrics & Analytics

What's your North Star metric? DAU dropped 15% — how do you investigate? Defines whether you're a data-driven PM or one who guesses.

Framework

North Star → Input Metrics → Guardrail Metrics → Breakdown by segment (geo, device, user type) → Hypothesis → Test

Always segment before concluding. A 15% DAU drop on iOS might be hiding a 50% drop on Android masked by iOS growth.

🧪

A/B Testing

Design an experiment for [feature]. Tests statistical thinking, experimentation maturity and ability to avoid common pitfalls.

Framework

Hypothesis → Primary metric + guardrails → Randomisation unit → Sample size → Duration → Decision framework → Ship or kill

The most common failure: running the test for 2 days and calling it significant. Always account for novelty effects and weekly cycles (run at least 1–2 full weeks).

🗂️

Prioritisation

Rank these 6 features for next quarter. Tests your ability to apply frameworks, handle trade-offs and say no with conviction.

Framework

Clarify goals → Apply RICE/ICE/MoSCoW → Estimate each dimension → Stack-rank → Address stakeholder pushback → Communicate what you're NOT building

The best prioritisation answer includes what you're explicitly NOT building and why. That's the sign of a PM who has actually shipped a roadmap.

🔎

Root Cause Analysis

A metric dropped — diagnose it. Tests structured thinking, data intuition and ability to stay calm when things go wrong.

Framework

Is it real (data issue)? → When did it start? → Seasonal? → Where (segment, geo, platform)? → What changed? → Hypothesis → Recommend fix

Always check for data pipeline issues first. 40% of 'metric drops' are actually logging bugs. Never escalate without ruling out instrumentation failures.

🤼

Stakeholder Management

CEO wants a feature added to your frozen roadmap. Engineering says it'll take 3× longer than sales promised. How do you handle it?

Framework

Listen to understand the underlying need → Data-back your position → Offer alternatives → Escalate only as last resort → Document decisions

The best PMs push back with data, not opinion. 'Based on our Q2 goal of improving checkout conversion, adding this now would cost us our primary metric' lands better than 'I don't think we should do this.'

What Separates Good PMs from Hired PMs

User Empathy

Can they identify a specific user pain — not just a feature request?

Structured Thinking

Do they clarify, segment, and prioritise before jumping to solutions?

Data-First Instinct

Is their first response 'I'd look at the data' not 'I think'?

Business Context

Do they connect product decisions to company revenue and strategy?

Decisive Prioritisation

Can they say no to good ideas — with reasoning that stands up?

Engineering Empathy

Do they respect constraints, ask about complexity, avoid scope creep?

Free Resources

SpeakWell AI is not affiliated with these websites. Links open in a new tab.

Practice 12 PM scenarios with an AI interviewer

Product design, metrics, A/B testing, root cause analysis, stakeholder management, and company-specific rounds at Amazon, Google and Flipkart — all with real-time AI coaching.