📐The PM Case Interview
Open-ended business cases that test how you structure ambiguity. Common at consulting-adjacent companies and senior PM loops.
Case interviews show up unexpectedly at senior PM loops. Without specific prep, you'll flounder; with it, you can demonstrate strategic thinking and business judgment.
A PM case interview is a 30-45 minute business case where you're given an ambiguous prompt and asked to structure the analysis, identify the key issues, and recommend a path. Less about being right; more about structured thinking + business savvy + crisp communication.
The format
Prompt examples:
- "Our new feature launched and adoption is flat. Diagnose and recommend."
- "We're considering acquiring a competitor for $100M. Walk me through the analysis."
- "Our key customer is threatening to leave. What do you do?"
You'll have 5-10 min to think (sometimes with paper, sometimes not), then 20-30 min to talk through your answer with the interviewer probing.
The structure
- Clarify (3 min). Don't dive in. Ask 3-5 clarifying questions to scope.
- Frame (3 min). "I'll think about this in three buckets: A, B, C." Set up your structure.
- Walk through (15-20 min). Bucket by bucket. State hypotheses, walk through implications, identify the most important.
- Recommendation (3 min). Pick a clear path. Justify.
- Risks + next steps (2 min). What could go wrong, what you'd do tomorrow.
What's tested
- Structured thinking. Did you frame the problem cleanly?
- Business savvy. Do you understand market dynamics, unit economics, competitive context?
- Quantitative judgment. Can you size, estimate, prioritize?
- Communication. Can you walk the interviewer through your reasoning?
- Synthesis. Can you pull threads into a clear recommendation?
Common cases
- Market entry: should we expand to [region/segment]?
- Pricing change: should we raise/lower pricing? Switch model?
- Product diagnosis: why is metric X declining?
- Acquisition: should we buy [target]?
- Strategic response: competitor launched X, what do we do?
Frameworks that help
- Porter's 5 Forces for market structure
- 3 C's (Company, Customers, Competition) for market entry
- Profitability framework (Revenue - Cost) for financial diagnosis
- STAR/SCQA for stakeholder narrative
Don't recite. Use as scaffolding.
What separates A from B
- A: Drives the structure. Frames the problem in their own words, doesn't wait for prompts.
- B: Reactive. Asks 'what should I look at?' constantly.
- A: Quantitative. Sizes the problem, prioritizes by impact.
- B: Qualitative-only.
- A: Clear recommendation. Picks a path with reasoning.
- B: Hedges. "It depends."
Practice
Practice with case-interview prep books (Case in Point by Cosentino is the canonical). 15-20 practice cases over 2 weeks builds fluency. Mock cases with peers help calibrate.
Real-world examples
Cases appear in senior PM loops at companies that recruit ex-consultants (Stripe, McKinsey-affiliated startups, certain VCs). Knowing the format and frameworks prevents being caught off-guard.
Go deeper — recommended reading
Interview questions (1)
Q1Our new feature launched 6 weeks ago. Adoption is 4%. Walk me through your analysis.casesenior▼
Clarify. What was the launch goal? Target adoption? Who's the target user? What does adoption mean specifically (used once vs used weekly)?
Frame. I'll structure around three buckets: (1) is the feature discoverable, (2) is the feature usable, (3) is the feature valuable.
Walk through.
Discoverability. Are users actually being exposed to the feature? Check entry points (in-product, email, push). If only 20% of eligible users have ever been exposed, that's the primary problem, not the feature itself.
Usability. Of users who see the feature, how many engage? If exposure is 60% but engagement is 5%, the feature has a discoverability-to-engagement gap — maybe UX is unclear, value prop weak.
Value. Of users who try, how many return? If first-use is high but D7 retention of feature is 5%, the feature works for first impression but doesn't deliver sustained value.
Diagnosis. Look at the funnel data; the bottleneck is usually obvious once sliced.
Recommendation. Suppose the bottleneck is usability — users see, try, but bounce. Three options: (a) redesign UX based on session replay analysis, (b) add in-product onboarding to the feature, (c) email education sequence for users who try once.
I'd recommend (a) + (b) — UX redesign anchored in session replay insights, plus contextual onboarding. (c) is supplementary.
Risks. Redesign could regress current 4% if poorly tested. Mitigate with phased rollout.
Next steps. This week: session replay analysis on 20 first-time users. Next 2 weeks: design v2 + onboarding spec. 4 weeks: A/B test redesigned flow.
Success criterion. Move adoption from 4% to 12% over 90 days. If we hit 8%, real progress; <6% means the feature itself might not have product-market fit and we should consider killing.