# 100 PM interview questions: frameworks for mid to executive levels

> A visual framework library and 100 model answers for product sense, strategy, metrics, growth, leadership, and AI, built for mid to VP Product interviews.

- Author: Kevin Dalvi
- Published: 2026-09-29
- Read time: 42 minutes
- Tags: product, interview, strategy, frameworks, leadership
- Canonical: https://www.kevindalvi.com/blog/pm-interview-frameworks

Mid to executive Product Management interviews rarely reward memorized frameworks announced out loud. They reward structured judgment: you make the invisible structure visible through crisp problem framing, quantitative instincts, qualitative customer insight, and technical fluency that earns engineering trust.

This guide is a framework library plus 100 interview questions and model answers spanning product sense, strategy, prioritization, metrics, growth, leadership, and AI. Use it as a workout, not a script. At VP level, the framework should almost disappear, you want to sound like you think structurally, not like you swallowed a PM textbook.

- **100** — Q&A drills
- **40+** — Frameworks
- **8** — Core mental models
- **7** — Interview clusters

### Chart: Signal weight in senior / VP product loops

Illustrative interview emphasis for mid→executive loops: strategy and leadership rise; pure feature brainstorming falls.

- Strategy: 22
- Leadership: 18
- Product sense: 16
- Metrics: 14
- Execution: 12
- Technical / AI: 18

## How to use this library

For each prompt, pick one primary framework and one backup. Narrate decisions, tradeoffs, and metrics. Bring technical depth: latency budgets, instrumentation, data models, AI failure modes, migration risk. Pair every quantitative claim with a qualitative customer truth.

### The portable senior PM OS

You do not need forty frameworks on cue. Keep these eight mentally available and compose them invisibly.

- Product design: Goal → User → Problem → Prioritize → Solution → Metrics
- Strategy: Diagnosis → Guiding Policy → Actions
- Prioritization: Impact → Confidence → Effort → Strategic Alignment
- Metrics: North Star → Inputs → Guardrails
- Diagnosis: Clarify → Segment → Funnel → Hypotheses → Validate
- Growth: Acquisition → Activation → Retention → Referral → Revenue
- Leadership: Context → Decision → Influence → Outcome → Learning
- AI: Problem → Capability → UX → Quality/Risk → Economics → Metrics

**Biggest interview mistake**

Announcing the framework and mechanically walking through it. At VP level, structure should be felt, not labeled. Sound like an operator making decisions, not a tutor grading a rubric.

### Flow: A reusable interview answer loop

Open-source Mermaid diagram: clarify → structure → decide → measure → learn.

```mermaid
flowchart LR
  A[Clarify goal] --> B[Name user / constraint]
  B --> C[Structure with a framework]
  C --> D[Decide with tradeoffs]
  D --> E[Define metrics]
  E --> F[State next experiment]
  F --> A
```

## Framework library by question type

Different questions test product sense, strategy, execution, analytics, leadership, prioritization, and technical judgment. Keep a library, not a single universal hammer.

### Framework: “Design a product for X.”

- **CIRCLES**: Comprehend situation → Identify customer → Report needs → Cut/prioritize → List solutions → Evaluate tradeoffs → Summarize.
- **User → Problem → Solution → Metrics**: Start with a specific user, identify their biggest problem, solve it, then define success.
- **Jobs to Be Done (JTBD)**: Understand the underlying “job” the customer is hiring the product to accomplish.
- **Design Thinking**: Empathize → Define → Ideate → Prototype → Test.

### Framework: “How would you improve X?”

- **Goals → Users → Pain → Solutions → Prioritize → Metrics**: Establish what improvement means before proposing features.
- **Customer Journey Map**: Walk discovery → onboarding → usage → retention to find friction.
- **HEART**: Evaluate Happiness, Engagement, Adoption, Retention and Task Success.

### Framework: “What is your favorite product? How would you improve it?”

- **Product → User → Value → Strengths → Gaps → Improvements**: Explain why the product works before jumping into what you would change.

### Framework: “Should we build X?”

- **Desirability / Viability / Feasibility**: Ask whether customers want it, whether it creates business value, and whether you can build it.
- **Opportunity Solution Tree**: Connect outcome → opportunities → solutions → experiments.
- **Build / Buy / Partner**: Compare developing internally against purchasing or partnering.

### Framework: “What should we build next?” / prioritization

- **RICE**: Score Reach × Impact × Confidence ÷ Effort.
- **ICE**: Score Impact × Confidence × Ease for lightweight prioritization.
- **Impact vs. Effort**: Plot opportunities by customer/business impact and implementation cost.
- **MoSCoW**: Must / Should / Could / Won’t-have features.
- **Kano Model**: Separate basics, performance features, and delights.
- **RICE + Strategic Alignment**: Combine quantitative scoring with company strategy.
- **Value vs. Complexity**: Favor high-value, lower-complexity work while naming tradeoffs.

### Framework: “What is the strategy for this product?”

- **Diagnosis → Guiding Policy → Actions**: Rumelt-style: understand the challenge, choose an approach, make coordinated choices.
- **Playing to Win**: Where will we play? How will we win? What capabilities and systems are required?
- **Product Strategy Stack**: Mission → Strategy → Product strategy → Roadmap → Goals.
- **Good Strategy / Bad Strategy**: Strategy requires explicit choices and tradeoffs, not a pile of goals.

### Framework: “How would you enter this market?”

- **Market → Customer → Competition → Differentiation → GTM**: Attractiveness, target segment, advantage, and route to market.
- **5Cs**: Company, Customers, Competitors, Collaborators, Context.
- **Porter's Five Forces**: Rivalry, entrants, substitutes, supplier power, buyer power.

### Framework: “How big is this opportunity?”

- **TAM / SAM / SOM**: Total market, addressable market, realistically obtainable market.
- **Top-down + Bottom-up**: Cross-check industry estimates against customers × price/usage.

### Framework: “Who are the customers?” / “What is the problem?”

- **Segmentation**: Divide by needs, behaviors, characteristics, or use cases.
- **ICP / Persona**: Profile most likely to derive significant value.
- **JTBD**: Segment around what customers are trying to accomplish.
- **5 Whys**: Move from symptoms to the underlying problem.
- **Problem Statement**: User + need + context + friction + desired outcome.

### Framework: “What is the MVP?” / “How would you launch?”

- **Hypothesis → Riskiest Assumption → Experiment**: Smallest build that tests the biggest uncertainty.
- **Walking Skeleton**: Thin end-to-end experience, not isolated fragments.
- **MVP / MLP**: Minimum viable vs. minimum lovable experience.
- **GTM Framework**: Customer → positioning → channels → pricing → launch → measurement.
- **Beachhead Strategy**: Win one narrow, high-value segment before expanding.
- **Positioning Framework**: Target, problem, category, differentiated benefit, alternatives.

### Framework: Growth, retention, metrics, diagnosis

- **AARRR / Pirate Metrics**: Acquisition → Activation → Retention → Referral → Revenue.
- **Growth Loops**: Usage itself generates additional users or usage.
- **North Star Metric**: Metric most representative of recurring customer value.
- **Input → Output → Outcome**: Separate activity from results and impact.
- **Guardrail Metrics**: Prevent optimizing one outcome from damaging another.
- **Clarify → Segment → Funnel → Hypotheses → Validate**: Structured metric-drop diagnosis.
- **MECE Issue Tree**: Mutually exclusive, collectively exhaustive causes.
- **Revenue Equation**: Users × conversion × frequency × price.

### Framework: Experiments, pricing, competition, roadmap

- **Hypothesis → Metric → Control/Treatment → Decision**: Causal hypothesis and success criteria before running.
- **Primary + Guardrail Metrics**: Judge intended improvement and unintended harm.
- **Value-Based Pricing**: Price to customer value, not cost-plus alone.
- **Good-Better-Best**: Differentiated packaging tiers.
- **Customer → Strategic Importance → Differentiate → Response**: Don’t copy competitors before validating customer care.
- **Outcomes → Opportunities → Initiatives**: Roadmap around outcomes, not a feature dump.
- **Now / Next / Later**: Direction without false precision on distant dates.

### Framework: Leadership, collaboration, executive judgment

- **STAR + Product Layer**: Situation → Task → Action → Result with product decisions and metrics.
- **STAR-L**: Add Learning after Result for failure stories.
- **Principles → Evidence → Tradeoffs → Decision**: Align around outcomes, not winning arguments.
- **Stakeholder → Incentives → Alignment → Influence → Outcome**: Influence without authority.
- **Disagree & Commit**: Argue with evidence, then support the decision.
- **Reversible vs. Irreversible**: Speed on two-way doors; rigor on one-way doors.
- **Learn → Align → Act**: 30/60/90 day plan structure.
- **Strategy → People → Operating System → Outcomes**: VP/Head of Product leadership model.

### Framework: AI product judgment

- **User Problem → AI Capability → UX → Reliability → Economics**: Start with the problem; justify AI cost and uncertainty.
- **Task → Autonomy → Tools → Risk → Human Control**: Decide what agents own vs. where humans stay in control.
- **Quality → Latency → Cost → Reliability → Safety → Adoption**: Evaluate AI products beyond classic product metrics.
- **Transparency → Control → Verification → Recovery**: Build trust when AI fails gracefully.

### Chart: Framework families to keep sharp

Coverage across the library, prioritization and metrics tools are used most often under time pressure.

- Product sense: 9
- Strategy: 11
- Prioritization: 10
- Metrics: 12
- Growth / GTM: 10
- Leadership: 10
- AI / technical: 8

## 100 questions and model answers

Each card includes level (mid / senior / executive), category tags, frameworks to lean on, a structured answer, plus quantitative and qualitative angles. Practice out loud with a timer: 3 to 5 minutes for mid prompts, 8 to 12 for executive strategy cases.

### Product sense (Q1 to Q12)

### Q1. Design a product that helps busy professionals manage personal finances.

Level: mid · Category: Product Sense
Frameworks: CIRCLES, JTBD, User → Problem → Solution → Metrics

Clarify the situation (time-poor W-2 professionals with fragmented accounts), pick one primary user, report jobs (see cash, avoid overdraft, pay down high-APR debt), cut scope to one beachhead job, list 2 to 3 solutions, evaluate tradeoffs (aggregation vs. coaching vs. autopay), and summarize with success metrics.

Quantitative: Target: 40% of activated users link ≥2 accounts in week 1; D30 retention ≥25% among users who completed first “money snapshot.”

Qualitative: Interview users who already use Mint/YNAB/spreadsheets; map the job “feel in control before payday” rather than “track every transaction.”

### Q2. How would you improve LinkedIn for job seekers?

Level: senior · Category: Product Sense
Frameworks: Goals → Users → Pain → Solutions → Metrics, Customer Journey Map, HEART

Define improvement as time-to-relevant interview, not feed engagement. Map journey from intent → profile → search → apply → response. Attack the biggest drop: opaque application status and weak matching. Propose ranked fixes with HEART metrics.

Quantitative: Cut median days from “Actively looking” to first recruiter reply by 20%; lift application→response rate from ~5% to 8% in a controlled geo.

Qualitative: Talk to 15 active seekers about where hope dies, usually silence after apply, not the search UI.

### Q3. What is your favorite product and how would you improve it?

Level: mid · Category: Product Sense
Frameworks: Product → User → Value → Strengths → Gaps → Improvements

Pick a product you use weekly. Explain the primary user and the core value loop before proposing changes. Name 2 strengths that create moat, 1 structural gap, then 1 to 2 improvements that reinforce the loop rather than decorate the UI.

Quantitative: Improvement hypothesis with a measurable proxy (e.g., +10% completion of the core action within 7 days).

Qualitative: Show taste: why the product works emotionally and operationally, not just a feature wishlist.

### Q4. Design an onboarding experience for a B2B analytics tool.

Level: senior · Category: Product Sense
Frameworks: Walking Skeleton, Activation, JTBD

Define activation as “first insight shared with a stakeholder,” not “account created.” Ship a walking skeleton: connect data → guided insight → share link. Defer customization until value is tasted.

Quantitative: Activation rate (insight shared / signup) ≥35% in 14 days; time-to-first-insight P50 < 20 minutes.

Qualitative: Shadow 5 analysts; note where they paste screenshots into Slack, that’s the real job.

### Q5. How would you raise the product bar for consumer quality at scale?

Level: executive · Category: Product Sense
Frameworks: Principles → Examples → Reviews → Coaching

Publish crisp quality principles (latency budgets, empty states, error recovery), show gold-standard examples, install review rituals (design + PM + eng), and coach managers to reject “works on my laptop” launches.

Quantitative: Track escaped defects / release, P95 interaction latency, and CSAT on top tasks; set quarterly bar lifts.

Qualitative: Make quality social: celebrate teams that cut support volume, not just those who ship volume.

### Q6. A competitor copied your core feature. What do you do?

Level: mid · Category: Product Sense
Frameworks: Customer → Strategic Importance → Differentiate → Response, Build / Ignore / Differentiate

Validate whether customers care and whether the copy changes win/loss rates. If table-stakes, match parity surgically; if differentiating, double down on the layer they can’t copy (workflow, data, trust, distribution).

Quantitative: Win-rate change in competitive deals ±3pts within 2 quarters; churn reason codes tagged “competitor feature.”

Qualitative: Ask lost deals: “What made their version good enough?” before staffing a clone project.

### Q7. Should we build an AI chatbot into our help center?

Level: senior · Category: Product Sense
Frameworks: Desirability / Viability / Feasibility, Problem → AI Capability → UX → Economics

Start from deflection and time-to-resolution, not chatbot novelty. Score desirability (users want faster answers), viability (support cost, CSAT), feasibility (retrieval quality on your docs). Pilot with human handoff and clear failure recovery.

Quantitative: Deflect ≥25% of L1 tickets with CSAT ≥ parity; cost/query under support-agent fully-loaded minute cost.

Qualitative: Trust requires citations, easy escalate-to-human, and visible when the model is unsure.

### Q8. How do you decide what not to build?

Level: mid · Category: Product Sense
Frameworks: Opportunity Solution Tree, Opportunity Cost, MoSCoW

Frame the desired outcome, list opportunities, and kill solutions that don’t change the outcome metric. Explicit Won’t-haves protect focus. Narrate opportunity cost in roadmap reviews.

Quantitative: Keep WIP ≤ N initiatives per squad; kill or park anything without a leading indicator move in 6 weeks.

Qualitative: Great PMs are remembered for what they deferred with conviction.

### Q9. Design a marketplace MVP for local services.

Level: senior · Category: Product Sense
Frameworks: Hypothesis → Riskiest Assumption → Experiment, Beachhead Strategy

Riskiest assumption is usually liquidity (supply + demand in a narrow geo/category), not payments UI. Beachhead: one city × one high-frequency service. Test matching with concierge before full automation.

Quantitative: Supply: ≥50 active pros; demand: ≥200 jobs/week; fill rate ≥60% before expanding category.

Qualitative: Talk to both sides weekly; marketplace strategy is interview volume dressed as product.

### Q10. How do you evaluate whether a 0→1 bet should become a product line?

Level: executive · Category: Product Sense
Frameworks: Retention + Pull + Growth + Economics, Strategic Fit

Look for retention that doesn’t need heroics, organic pull (waitlists, inbound), a path to growth loops, and unit economics that improve with scale. Require strategic fit with distribution and brand.

Quantitative: D30 retention cohort stable 2+ months; payback < target CAC months; ≥30% of new users from non-paid loops.

Qualitative: Kill charming demos that only work with founder-led onboarding.

### Q11. Walk me through how you’d redesign checkout.

Level: mid · Category: Product Sense
Frameworks: Funnel Analysis, HEART, Guardrails

Instrument step conversion, segment by device/geo/payment method, fix the largest drop first. Guardrail: fraud rate and support contacts must not spike.

Quantitative: Checkout completion +8% absolute; payment-error rate −30%; fraud rate within ±0.1pp.

Qualitative: Watch session replays for hesitation, not just click maps.

### Q12. How would you use AI to improve search relevance?

Level: senior · Category: Product Sense
Frameworks: Problem → AI Capability → UX → Quality/Risk → Metrics

Define relevance failure modes (zero results, wrong intent). Use embeddings/rerankers where they beat lexical baselines. Expose why results appeared; measure NDCG/CTR and task success, plus latency/cost budgets.

Quantitative: Lift query→click@3 by 12%; keep P95 latency < +50ms; cost/query within SLO.

Qualitative: Pair online A/B with offline judgment sets reviewed by domain experts.

### Flow: Product design path (invisible CIRCLES)

Use the spine; don’t recite the acronym unless asked.

```mermaid
flowchart TD
  G[Goal] --> U[User]
  U --> P[Problem]
  P --> C[Cut scope]
  C --> S[Solutions]
  S --> T[Tradeoffs]
  T --> M[Metrics]
```

### Strategy and markets (Q13 to Q28)

### Q13. What is the strategy for this product over the next 3 years?

Level: executive · Category: Strategy
Frameworks: Diagnosis → Guiding Policy → Actions, Playing to Win

Diagnose the binding constraint (distribution, switching costs, category confusion). State a guiding policy that forces tradeoffs. List coordinated actions across product, GTM, and ops, not a feature laundry list.

Quantitative: Tie strategy to 2 to 3 outcome metrics (e.g., % revenue from ICP, NRR, share in beachhead segment).

Qualitative: Bad strategy sounds like “be the best.” Good strategy names what you will stop doing.

### Q14. How would you enter the SMB CRM market against incumbents?

Level: senior · Category: Strategy
Frameworks: 5Cs, Beachhead, Differentiation → GTM

Narrow beachhead (e.g., agencies with 5 to 20 seats). Differentiate on workflow depth, not horizontal CRM completeness. GTM via communities/partners where Salesforce is overkill.

Quantitative: SOM year-1: $8 to 12M ARR from beachhead; logo win-rate ≥25% in ICP opportunities.

Qualitative: Win stories should sound like “finally fits how we sell,” not “cheaper Salesforce.”

### Q15. How big is the opportunity for AI meeting notes?

Level: mid · Category: Strategy
Frameworks: TAM / SAM / SOM, Top-down + Bottom-up

Top-down: knowledge workers × willingness to pay for meeting tooling. Bottom-up: ICP companies × seats × ARPU × adoption. Cross-check; be honest about substitutes (human notes, Zoom summaries).

Quantitative: Show TAM/SAM/SOM with assumptions table; sensitivity on ARPU ±30% and adoption ±10pp.

Qualitative: Call out the real competitor: “I’ll remember” and existing calendars.

### Q16. Where will we play and how will we win?

Level: executive · Category: Strategy
Frameworks: Playing to Win, Capabilities & Systems

Choose arenas (segment, geo, job-to-be-done) and a winning logic (cost, experience, ecosystem, data). Name required capabilities and the management systems that reinforce them.

Quantitative: Portfolio: ≥70% eng capacity mapped to chosen arenas; kill orphan initiatives quarterly.

Qualitative: If everything is a priority arena, you don’t have a strategy.

### Q17. Should we expand internationally next year?

Level: senior · Category: Strategy
Frameworks: Market → Customer → Competition → GTM, Opportunity Cost

Score markets on demand signal, payment/ops readiness, localization cost, and competitive intensity. Expand only if beachhead PMF is durable and support can absorb load.

Quantitative: Gate: home-market NRR ≥110%, support SLAs green, localization payback model < 18 months.

Qualitative: Premature multi-geo is a tax on learning velocity.

### Q18. How do you analyze Porter’s Five Forces for our category?

Level: mid · Category: Strategy
Frameworks: Porter's Five Forces

Map rivalry, new entrants, substitutes, supplier power (cloud/model providers), buyer power. Translate into product bets: switching costs, unique data, partner leverage.

Quantitative: Quantify buyer power via discounting pressure and sales-cycle length trends.

Qualitative: Strategy implication > academic completeness.

### Q19. How do you allocate resources across a multi-product portfolio?

Level: executive · Category: Strategy
Frameworks: Strategic Importance × Opportunity × Evidence × Investment

Score products on strategic fit, market opportunity, evidence of traction, and capital intensity. Fund growth engines, starve zombies, and ring-fence platform investments that unlock multiple products.

Quantitative: Publish capacity allocation % by product each quarter; require kill/continue reviews for bottom quartile.

Qualitative: Allocation is strategy made visible, politics thrives in opacity.

### Q20. When do you kill a product?

Level: senior · Category: Strategy
Frameworks: Strategic Fit → Value → Traction → Economics → Opportunity Cost

Kill when strategic fit decays, customer value is niche, traction stalls despite investment, economics don’t improve, and opportunity cost is clear. Communicate with dignity and migration paths.

Quantitative: Predefine kill criteria (e.g., <X WAU after Y months at Z investment).

Qualitative: Killing well builds more trust than endless soft sunsets.

### Q21. How would you position against a free competitor?

Level: mid · Category: Strategy
Frameworks: Positioning Framework, Value Metric

Position on a job free tools can’t finish (compliance, collaboration, reliability, admin controls). Price on value metric buyers already budget for.

Quantitative: Conversion free→paid ≥4%; paid churn < free churn; win themes coded vs free.

Qualitative: Never compete only on “more features than free.”

### Q22. What does good strategy look like vs bad strategy here?

Level: executive · Category: Strategy
Frameworks: Good Strategy / Bad Strategy

Good: coherent diagnosis, guiding policy, coordinated actions. Bad: buzzword goals, fluff vision, laundry-list OKRs without tradeoffs.

Quantitative: Audit last planning deck: count true tradeoffs vs aspirational goals.

Qualitative: If a competitor could copy-paste your strategy slide, it isn’t strategy.

### Q23. How do you connect mission to roadmap?

Level: senior · Category: Strategy
Frameworks: Product Strategy Stack

Mission → company strategy → product strategy → roadmap themes → goals. Every roadmap item should trace upward; orphan features get cut.

Quantitative: % of roadmap items with explicit strategy link in planning tool ≥90%.

Qualitative: Engineers should be able to explain why this quarter’s work matters.

### Q24. Build, buy, or partner for identity/auth?

Level: mid · Category: Strategy
Frameworks: Build / Buy / Partner, TCO × Speed × Risk

Auth is rarely differentiating, buy/partner unless compliance forces control. Build only the thin layer that creates product-specific value.

Quantitative: TCO 3-year model + time-to-SOC2; choose option with lower risk-adjusted cost and faster GTM.

Qualitative: Differentiate on product experience, not reinventing OAuth.

### Q25. How should AI change our competitive strategy?

Level: executive · Category: Strategy
Frameworks: Problem → AI Capability → Economics, Differentiation

AI is a capability, not a strategy. Identify jobs where prediction/generation compounds your data advantage. Avoid me-too chat wrappers; invest where switching costs and workflow depth increase.

Quantitative: Track AI-attached revenue, gross margin impact, and differentiation win-rate themes.

Qualitative: Strategy question: what becomes 10× better because of our proprietary context?

### Q26. How do you define product-market fit rigorously?

Level: senior · Category: Strategy
Frameworks: Retention + Pull + Growth + Economics

Sustained retention, strong pull (organic demand), efficient growth path, and improving economics. Qualitative: users would be “very disappointed” without you for the core job.

Quantitative: Sean Ellis ≥40% as a signal, plus cohort retention curves flattening above baseline; CAC payback trending down.

Qualitative: PMF is a zone you maintain, not a trophy you hang.

### Q27. Who is the ICP and who is not?

Level: mid · Category: Strategy
Frameworks: ICP / Persona, Segmentation

Define firmographics/behaviors of high-LTV, low-CAC customers. Explicitly name anti-ICP that burns support and dilutes roadmap.

Quantitative: ICP revenue ≥60% of new ARR; sales discounting lower in ICP vs non-ICP.

Qualitative: Courage is turning down revenue that teaches the wrong product lessons.

### Q28. How do you influence the exec team on a portfolio bet?

Level: executive · Category: Strategy
Frameworks: Context → Evidence → Options → Tradeoffs → Recommendation

Frame in company outcomes, show evidence, present 2 to 3 options with tradeoffs, make a clear recommendation and ask. Bring kill criteria up front.

Quantitative: Decision latency from first memo to funded bet; revisit at pre-agreed checkpoints with data.

Qualitative: Executives buy clarity and risk management, not feature enthusiasm.

### Chart: Illustrative beachhead → expansion revenue

Win a narrow SOM first; expansion after liquidity/PMF signals, not before.

- Y0: 2
- Y1: 5
- Y2: 11
- Y3: 22
- Y4: 38

### Prioritization and execution (Q29 to Q45)

### Q29. How would you prioritize these 8 feature requests?

Level: mid · Category: Prioritization
Frameworks: RICE, Strategic Alignment

Score Reach/Impact/Confidence/Effort, then overlay strategic themes. Call out confidence gaps and what experiment would raise confidence cheaply.

Quantitative: Show ranked table with RICE scores and a sensitivity note if confidence is low.

Qualitative: Invite eng/design to challenge Effort and Impact assumptions live.

### Q30. RICE says A wins, but sales is screaming for B. What do you do?

Level: senior · Category: Prioritization
Frameworks: Goals → Interests → Decision Rights, Value vs. Complexity

Separate positions from interests. Quantify revenue at risk vs. long-term metric impact. Escalate with decision rights clear: PM owns roadmap unless exec overrides with explicit tradeoff.

Quantitative: If override: document expected revenue and review in 60 days; if miss, restore capacity.

Qualitative: Influence without contempt, sales pain is data.

### Q31. Explain ICE vs RICE and when you’d use each.

Level: mid · Category: Prioritization
Frameworks: ICE, RICE

ICE is fast for early teams; RICE adds Reach and divides by Effort for clearer tradeoffs. Use ICE in brainstorms; RICE when capacity is scarce and stakeholders need transparency.

Quantitative: Timebox scoring sessions; avoid false precision (Confidence to nearest 10%).

Qualitative: The framework is a conversation scaffold, not a truth machine.

### Q32. How do you balance tech debt vs features?

Level: senior · Category: Prioritization
Frameworks: Impact × Risk × Urgency

Quantify debt via incident rate, cycle time drag, and eng happiness. Schedule debt when it gates customer outcomes or compounds risk; avoid abstract “20% debt tax” without targets.

Quantitative: Reserve capacity tied to reliability SLOs; burn-down critical debt with dates.

Qualitative: Debt work needs customer language: “checkout fails less,” not “refactor module.”

### Q33. Engineering says this takes 6 months. How do you respond?

Level: executive · Category: Prioritization
Frameworks: Outcome → Constraints → Decomposition → Alternatives

Reconfirm the outcome, understand constraints, slice a walking skeleton or scope cut that learns in weeks. Explore buy/partner. Don’t accept binary 0 or 6 months without options.

Quantitative: Demand a 2-week spike with milestone risks; decide with better Effort confidence.

Qualitative: Senior PMs buy options on uncertainty, not waterfall certainty.

### Q34. How do you use the Kano model in practice?

Level: mid · Category: Prioritization
Frameworks: Kano Model

Survey/basics vs delighters; ensure Must-be quality before Performance features; use Delighters sparingly for differentiation.

Quantitative: Track basic defects (Must-be failures) separately from NPS-driving delights.

Qualitative: Shipping delighters on a buggy basic experience destroys trust.

### Q35. Build a Now / Next / Later roadmap for Q2.

Level: senior · Category: Prioritization
Frameworks: Now / Next / Later, Outcomes → Opportunities → Initiatives

Now: committed outcomes with evidence. Next: high-confidence bets. Later: directional themes. Avoid dated promises beyond confidence horizon.

Quantitative: Each Now item has owner, metric, and review date.

Qualitative: Later is strategy communication, not a backlog graveyard.

### Q36. What is MoSCoW and when is it useful?

Level: mid · Category: Prioritization
Frameworks: MoSCoW

Categorize Must/Should/Could/Won’t for a release train. Useful for launches with hard constraints (date, compliance). Weak for continuous discovery without outcome framing.

Quantitative: Must-haves ≤ capacity; if everything is Must, renegotiate date or scope.

Qualitative: Won’t list is a gift to stakeholders, it creates clarity.

### Q37. How do you create an operating system for product teams?

Level: executive · Category: Execution
Frameworks: Context → Standards → Coaching → Accountability

Shared strategy context, decision standards (PRDs, metrics, reviews), coaching for PMs, and accountability on outcomes, not story points.

Quantitative: Quarterly outcome hit-rate; decision quality reviews; time-to-insight on bets.

Qualitative: Process exists to improve decisions, not to generate artifacts.

### Q38. How do you run discovery without slowing delivery?

Level: senior · Category: Execution
Frameworks: Opportunity Solution Tree, Dual-track

Dual-track: small continuous discovery alongside delivery. Spike riskiest assumptions early. Keep discovery artifacts lightweight.

Quantitative: % of shipped work with validated assumption notes; kill rate of ideas pre-build.

Qualitative: Discovery theater (endless interviews, no decisions) is as bad as zero research.

### Q39. What does a good PRD look like at senior level?

Level: mid · Category: Execution
Frameworks: Problem → Outcomes → Non-goals → Metrics → Risks

Problem, users, outcomes, non-goals, success metrics, constraints, open questions, rollout. Short. Decision-oriented. Links to research and designs.

Quantitative: PRD length ≠ quality; measure whether eng/design can execute without Slack archaeology.

Qualitative: Write to create alignment, not to impress with thoroughness.

### Q40. How do you manage dependencies across teams?

Level: senior · Category: Execution
Frameworks: Decision Rights, RACI-lite, Milestones

Map critical path, assign single-threaded owners, define interfaces/contracts early, and escalate blockers with options, not complaints.

Quantitative: Dependency SLA: response <48h; milestone burn-up visible.

Qualitative: Most “cross-team failures” are unclear ownership dressed as process issues.

### Q41. How do you handle scope creep mid-sprint?

Level: mid · Category: Execution
Frameworks: Goals → Interests → Tradeoffs

Reconnect to the outcome. Trade explicitly: new scope in means something out or date moves. Document the decision.

Quantitative: Track change requests accepted vs deferred; protect activation metrics of the original bet.

Qualitative: Saying yes to everything is how roadmaps become fiction.

### Q42. How do you know a team is high-performing?

Level: executive · Category: Execution
Frameworks: Outcomes, Learning Velocity, Quality

Outcome delivery, learning velocity (cheap killed ideas), quality/reliability, healthy conflict, and stakeholder trust. Output vanity is insufficient.

Quantitative: Outcome OKRs, incident trends, eNPS/PM engagement, cycle time for validated learnings.

Qualitative: High performers make the org smarter every quarter.

### Q43. Walk through launching a major feature.

Level: senior · Category: Execution
Frameworks: GTM Framework, Primary + Guardrails

Beta with ICP, instrumentation, support runbooks, pricing/packaging if needed, staged rollout, success/guardrail metrics, and comms plan.

Quantitative: Rollout gates: error budget, adoption threshold, support ticket rate.

Qualitative: Launch is a product system, not a blog post day.

### Q44. How do you partner with design effectively?

Level: mid · Category: Execution
Frameworks: Problem → Insight → Exploration → Validation

Bring the problem and constraints, not pixel specs. Explore multiple directions, validate with users, decide together on tradeoffs.

Quantitative: Track design↔PM decision turnaround; reduce late-stage rework rate.

Qualitative: Handing design a solution brief wastes both crafts.

### Q45. How do you partner with engineering effectively?

Level: senior · Category: Execution
Frameworks: Why / What / How

Own clarity on why/what and success metrics; co-own feasibility and sequencing; respect eng ownership of how. Invite technical alternatives early.

Quantitative: Estimate accuracy improves when spikes are funded; track rework from unclear requirements.

Qualitative: Best PMs make engineers faster by reducing thrash.

### Impact × Confidence × Effort × Alignment

When stakeholders collide, put numbers and strategy on the same table. Low confidence is a call for a cheaper experiment, not a louder opinion.

- Raise Confidence with spikes, prototypes, or customer evidence
- Re-score after new information, don’t marry the first ranking
- Document overrides so politics becomes an explicit tradeoff

### Metrics and diagnosis (Q46 to Q62)

### Q46. What should the North Star metric be for a collaboration product?

Level: senior · Category: Metrics
Frameworks: North Star, Value → Frequency → Breadth

Pick a metric that captures recurring value, e.g., weekly docs edited by ≥2 people, not vanity signups. Validate it moves with retention and revenue.

Quantitative: Correlation of NSM with paid retention; leading indicators dashboarded under NSM.

Qualitative: If NSM can be gamed without value, redesign it.

### Q47. A key metric dropped 20%. What do you do?

Level: mid · Category: Metrics
Frameworks: Clarify → Segment → Funnel → Hypotheses → Validate

Verify definition/instrumentation, segment (platform, geo, cohort, channel), localize funnel stage, generate MECE hypotheses, validate with data/experiments. Communicate confidence levels.

Quantitative: Time-to-root-cause; % of “metric drops” that are instrumentation vs real.

Qualitative: Calm structure beats panic feature shipping.

### Q48. Conversion suddenly dropped. Diagnose it.

Level: senior · Category: Metrics
Frameworks: Funnel Analysis, Internal vs. External

Find the broken step, then segment. Check releases, tracking, seasonality, competitors, payment processors. Reproduce on affected surfaces.

Quantitative: Attribute drop to step × segment with statistical significance; ship fix or rollback with measured recovery.

Qualitative: Always ask: did the world change or did we change the world?

### Q49. Revenue is down 12% QoQ. How do you diagnose?

Level: executive · Category: Metrics
Frameworks: Revenue Equation, Funnel × Segmentation

Decompose: users × conversion × frequency × price × mix. Separate new vs expansion vs churn. Localize before prescribing strategy.

Quantitative: Bridge chart from prior quarter to current with each driver’s contribution.

Qualitative: Exec rooms need bridges, not vibes.

### Q50. Engagement is down. What’s your tree?

Level: mid · Category: Metrics
Frameworks: Engagement Tree

Break into active users × sessions/user × depth/duration. Find which factor moved. Hypothesize product, seasonality, or acquisition mix shifts.

Quantitative: Quantify each node’s contribution to the engagement decline.

Qualitative: “Engagement” without a tree is a slogan.

### Q51. How do you choose guardrail metrics for an experiment?

Level: senior · Category: Metrics
Frameworks: Primary + Guardrails

Pick guardrails that catch likely harm: retention, latency, refunds, trust/CSAT, margin. Keep few enough that teams can decide.

Quantitative: Pre-register guardrail thresholds; auto-pause on severe breaches.

Qualitative: Optimizing conversion while torching trust is how you win the wrong game.

### Q52. Explain Input vs Output vs Outcome metrics.

Level: mid · Category: Metrics
Frameworks: Input → Output → Outcome

Inputs (interviews run), outputs (features shipped), outcomes (retention, revenue, task success). Manage to outcomes; use inputs as leading process health.

Quantitative: OKRs weighted to outcomes; outputs as milestones only.

Qualitative: Shipping is not succeeding.

### Q53. How would you run an A/B test for pricing page copy?

Level: senior · Category: Metrics
Frameworks: Hypothesis → Metric → Control/Treatment → Decision

Hypothesis: clearer value prop lifts trial starts without hurting qualified pipeline. Primary: trial start rate; guardrails: sales-accepted opportunity rate, refunds.

Quantitative: MDE and sample size calculated; run full business cycle if B2B.

Qualitative: Stat-sig vanity with worse lead quality is a failed test.

### Q54. How do you prevent metric theater in the org?

Level: executive · Category: Metrics
Frameworks: North Star, Guardrails, Reviews

Fewer metrics, clear owners, instrumentation audits, and reviews that ask “what did we learn/kill?” Celebrate honest misses.

Quantitative: Dashboard count ↓; decision docs citing metrics ↑; audit false positives quarterly.

Qualitative: Culture eats dashboards for breakfast.

### Q55. What is HEART and when do you use it?

Level: mid · Category: Metrics
Frameworks: HEART

Happiness, Engagement, Adoption, Retention, Task success, useful for UX-heavy products where a single funnel metric is incomplete.

Quantitative: Pick 1 to 2 HEART dimensions per initiative to avoid dashboard sprawl.

Qualitative: HEART without task success often becomes vibe metrics.

### Q56. Users aren’t retaining. What do you do?

Level: senior · Category: Metrics
Frameworks: Acquisition → Activation → Engagement → Retention, Cohort Analysis

Diagnose whether you acquire the wrong users, fail activation, or fail ongoing value. Cohort by channel and first-week behavior. Fix the earliest broken stage.

Quantitative: Activation→D30 conversion lift; channel quality scores.

Qualitative: Retention problems are often activation problems wearing a trench coat.

### Q57. How do you use cohort analysis in an interview?

Level: mid · Category: Metrics
Frameworks: Cohort Analysis

Compare retention by signup week/channel/persona. Look for improving or degrading curves after product changes. Narrate causality cautiously.

Quantitative: Show a simplified cohort table and one insight you’d act on.

Qualitative: Interviewers want structured thinking more than perfect SQL.

### Q58. Design a metrics system for an AI feature.

Level: senior · Category: Metrics
Frameworks: Quality → Latency → Cost → Reliability → Safety → Adoption

Track model quality (human eval + online), latency, unit cost, failure/escalation rates, safety incidents, and adoption/retention of the AI surface.

Quantitative: Cost per successful task; % tasks needing human correction; adoption among ICP.

Qualitative: AI metrics are product metrics plus model economics.

### Q59. Which metrics should board vs product team see?

Level: executive · Category: Metrics
Frameworks: North Star, Input → Output → Outcome

Board: outcomes and strategic health (growth, NRR, burn multiple). Product: diagnostic trees and experiment readouts. Don’t dump operational noise upward.

Quantitative: Board pack ≤10 metrics with bridges; product has drill-downs.

Qualitative: Altitude matching is an underrated leadership skill.

### Q60. How do you know an experiment succeeded?

Level: mid · Category: Metrics
Frameworks: Primary + Guardrails

Primary metric hit pre-registered threshold; guardrails held; result durable across key segments; decision documented (ship/iterate/kill).

Quantitative: Pre-commit success criteria; avoid peeking-driven decisions.

Qualitative: Success is a decision quality, not a green dashboard.

### Q61. How would you measure marketplace health?

Level: senior · Category: Metrics
Frameworks: Liquidity Metrics, AARRR

Fill rate, time-to-match, concentration risk, cohort retention both sides, take rate vs satisfaction. Watch chicken-egg by geo/category.

Quantitative: Liquidity scorecard weekly; alert on fill-rate drops >X%.

Qualitative: Marketplace metrics must be two-sided or they lie.

### Q62. What’s wrong with optimizing solely for DAU?

Level: mid · Category: Metrics
Frameworks: Guardrails, North Star

DAU can rise via spammy notifications while value and retention fall. Prefer value-tied NSM with guardrails on unsubscribes and session quality.

Quantitative: Correlate DAU changes with retention and complaint rates.

Qualitative: DAU is a flashlight, not a destination.

### Flow: Metric drop diagnosis

Clarify instrumentation before you rewrite the roadmap.

```mermaid
flowchart TD
  C[Clarify definition] --> S[Segment]
  S --> F[Funnel stage]
  F --> H[Hypotheses MECE]
  H --> V[Validate]
  V --> D{Real change?}
  D -->|No| I[Fix instrumentation]
  D -->|Yes| X[Ship fix / experiment]
```

### Growth, GTM, and pricing (Q63 to Q78)

### Q63. How would you grow this product from $2M to $10M ARR?

Level: senior · Category: Growth
Frameworks: AARRR, Growth Loops

Diagnose constraining stage. Build loops (viral, content, sales-assist, usage-based expansion). Pair product-led motions with targeted sales for enterprise.

Quantitative: Stage conversion lifts; expansion revenue %; loop contribution to new ARR.

Qualitative: Growth plans without a bottleneck diagnosis are fan fiction.

### Q64. Explain growth loops vs funnels.

Level: mid · Category: Growth
Frameworks: Growth Loops, Growth Funnel

Funnels are linear conversion; loops compound when output feeds input. Prefer loops for durable growth; use funnels to find leaks.

Quantitative: Measure loop factor (new users generated per active user per period).

Qualitative: Most “growth ideas” are funnel patches; few are true loops.

### Q65. Design a beachhead GTM for a new vertical.

Level: senior · Category: Growth
Frameworks: Beachhead, GTM Framework

Pick narrow ICP, sharp positioning, 1 to 2 channels, reference customers, playbooks for sales/CS, and exit criteria before adjacent expansion.

Quantitative: Win-rate, sales cycle, logo references, payback in beachhead before expansion.

Qualitative: Land-and-expand only works if land actually lands.

### Q66. How do you balance PLG and enterprise sales?

Level: executive · Category: Growth
Frameworks: Segmentation, Packaging, Motion Fit

PLG for land in SMB/mid-market; enterprise for security, procurement, multi-team expansion. Align packaging and product surfaces to motions without fragmenting the roadmap.

Quantitative: Self-serve ARR vs sales-assist ARR; attach rate of enterprise features.

Qualitative: Two motions need one product strategy or you get two mediocre products.

### Q67. How would you price a new Pro tier?

Level: mid · Category: Pricing
Frameworks: Value-Based Pricing, Willingness to Pay, Good-Better-Best

Map value metric, research WTP by segment, package Good-Better-Best with clear fences, test price locally or with sales assists before global rollout.

Quantitative: ARPU, conversion, expansion, and churn by cohort post-price change.

Qualitative: Cost-plus pricing leaves money and positioning on the table.

### Q68. Should we move to usage-based pricing?

Level: senior · Category: Pricing
Frameworks: Value Metric, Economics, Predictability

If value scales with usage and customers accept variable bills, yes, with caps/credits for predictability. Model margin, forecasting, and sales compensation impacts.

Quantitative: Gross margin, revenue volatility, customer bill shock tickets, NRR.

Qualitative: Usage pricing fails when the meter doesn’t match perceived value.

### Q69. How do you improve activation?

Level: mid · Category: Growth
Frameworks: Activation Definition, Time-to-Value

Define activation event tied to value, instrument paths, remove steps, use templates/defaults, trigger human assist for high-intent users.

Quantitative: Activation rate + time-to-activation P50/P90.

Qualitative: Activation is a product design problem before it is an email problem.

### Q70. Referral is flat. How do you approach it?

Level: senior · Category: Growth
Frameworks: AARRR, Growth Loops

Check if product is share-worthy at moments of delight; reduce friction; incentivize both sides carefully; ensure referred users activate. Don’t bolt on referral to a weak core.

Quantitative: Invite→signup→activation rates; K-factor; fraud/abuse rates.

Qualitative: Referrals amplify product love; they rarely create it.

### Q71. How do you launch in a category with noisy AI hype?

Level: executive · Category: GTM
Frameworks: Positioning, Trust, Proof

Position on a concrete job and measurable outcomes. Lead with proof (evals, case studies, security). Avoid hype vocabulary that invites skepticism.

Quantitative: Pipeline quality, sales-cycle length, win themes vs “AI theater” competitors.

Qualitative: In hype markets, trust is the differentiator.

### Q72. What’s your launch checklist?

Level: mid · Category: GTM
Frameworks: GTM Framework

Audience, message, channels, pricing, enablement, support, instrumentation, rollout plan, success/guardrails, postmortem date.

Quantitative: Launch scorecard reviewed at T+7/T+30.

Qualitative: If support learns about launch on Twitter, you failed operationally.

### Q73. How would you design a habit loop for a learning product?

Level: senior · Category: Growth
Frameworks: Habit Loop

Trigger (calendar/streak), action (5-min lesson), reward (progress/social proof), investment (personalized path). Align with real learning outcomes, not empty streaks.

Quantitative: D1/D7/D30 lesson completion; learning outcome assessments.

Qualitative: Dark-pattern streaks destroy long-term trust in edtech.

### Q74. How do you handle a pricing revolt after a change?

Level: executive · Category: Pricing
Frameworks: Segmentation, Communication, Grandfathering

Segment impact, communicate value clearly, offer grandfathering/migration paths for loyal cohorts, and fix true value gaps. Don’t panic-revert without learning.

Quantitative: Churn, expansion, ticket volume, sentiment by segment post-change.

Qualitative: Pricing is a product change, treat it with the same rigor as a major launch.

### Q75. How do you find the biggest funnel constraint?

Level: mid · Category: Growth
Frameworks: Growth Funnel

Quantify conversion at each stage; estimate impact of lifting each; pick the stage with highest expected value given effort. Re-measure after interventions.

Quantitative: Expected ARR impact table per stage lift.

Qualitative: Optimizing a non-bottleneck is motion without progress.

### Q76. How do you create positioning that sales will actually use?

Level: senior · Category: GTM
Frameworks: Positioning Framework

Co-create with sales using win/loss; one sentence competitive alternative; proof points; talk tracks. Update from call recordings quarterly.

Quantitative: Win-rate and message adherence in call scoring.

Qualitative: Positioning that lives only in a Notion doc is decoration.

### Q77. When should you turn on paid acquisition?

Level: executive · Category: Growth
Frameworks: Unit Economics, PMF Signals

After organic retention/activation look healthy and payback math works at target scale. Start with small tests; scale channels with proven CAC/LTV.

Quantitative: Payback period, LTV:CAC, channel-level retention quality.

Qualitative: Paid is rocket fuel, useless or explosive without a working engine.

### Q78. Explain Good-Better-Best packaging.

Level: mid · Category: Pricing
Frameworks: Good-Better-Best

Three tiers with clear fences: Good solves core job; Better adds collaboration/admin; Best adds advanced controls/scale. Avoid overlapping SKUs.

Quantitative: Attach rates by tier; upgrade paths; sales discounting by tier.

Qualitative: If buyers can’t tell tiers apart in 10 seconds, packaging failed.

### Chart: Pirate metrics: find the constraint

Example conversion index (relative). Attack the weakest stage first.

- Acquisition: 72
- Activation: 41
- Retention: 55
- Referral: 28
- Revenue: 47

### Leadership and behavioral (Q79 to Q90)

### Q79. Tell me about a product you launched.

Level: senior · Category: Leadership
Frameworks: STAR + Product Layer

Situation/context, your ownership, key product decisions (what you cut), cross-functional actions, and measurable results. Include technical constraints you respected.

Quantitative: Ship metrics, adoption, revenue/retention impact, and what you’d redo.

Qualitative: Interviewers listen for judgment under constraint, not hero narratives.

### Q80. Tell me about a failure.

Level: senior · Category: Leadership
Frameworks: STAR-L

Own a real miss, what you did, the result, and the learning you institutionalized (process/metric/principle). Avoid fake failures (“I worked too hard”).

Quantitative: Show the learning became a system change, not just a regret.

Qualitative: Credibility comes from scar tissue with upgrades.

### Q81. Tell me about a conflict with a stakeholder.

Level: mid · Category: Leadership
Frameworks: STAR, Goals → Interests → Tradeoffs

Specific disagreement, interests underneath, how you created a decision path, and outcome. Stay respectful and concrete.

Quantitative: Prefer stories with a measurable resolution.

Qualitative: Conflict stories without decisions sound like therapy.

### Q82. Tell me about disagreeing with leadership.

Level: executive · Category: Leadership
Frameworks: Disagree & Commit

You raised a clear, evidence-based dissent, escalated appropriately, then committed fully to the decision, including helping it succeed.

Quantitative: Show outcome and relationship trust after the disagree.

Qualitative: Chronic disagree-without-commit is organizational sand.

### Q83. How do you influence without authority?

Level: senior · Category: Leadership
Frameworks: Stakeholder → Incentives → Alignment → Influence → Outcome

Map stakeholders and incentives, create shared metrics, use prototypes/data, build coalitions, and make the easy path the aligned path.

Quantitative: Cite a concrete win where you had no direct reports over the decision.

Qualitative: Authority is optional; clarity and empathy are not.

### Q84. What makes a great VP/Head of Product?

Level: executive · Category: Leadership
Frameworks: Strategy → People → OS → Outcomes

Set strategy, hire/coach PMs, build operating cadence, and own portfolio outcomes. Shift from being the best IC to building a decision factory.

Quantitative: Team outcome delivery, PM quality bar, exec trust scores.

Qualitative: If you’re still the bottleneck on every PRD, you’re not leading yet.

### Q85. How do you manage and coach PMs?

Level: senior · Category: Leadership
Frameworks: Context → Standards → Coaching → Accountability

Give strategy context, clear standards for quality, regular coaching on decisions, and accountability for outcomes. Differentiate support vs performance management.

Quantitative: PM ramp time; decision quality reviews; retention of top PMs.

Qualitative: Coaching is about better judgment under uncertainty.

### Q86. How do you handle ambiguity on day one of a new problem?

Level: mid · Category: Leadership
Frameworks: Knowns / Unknowns / Assumptions, 70% Rule

List knowns/unknowns/assumptions, identify riskiest unknown, pick a reversible next step. Act at ~70% information when stakes allow.

Quantitative: Time-to-first validated learning; assumption backlog burned down.

Qualitative: Ambiguity tolerance without structure is just wandering.

### Q87. What would you do in your first 30/60/90 days?

Level: executive · Category: Leadership
Frameworks: Learn → Align → Act

30: listen, map systems/metrics/customers. 60: align on problems and priorities. 90: ship meaningful momentum with measurable outcomes and a clear operating rhythm.

Quantitative: Define 90-day outcome targets with your manager in week 1.

Qualitative: Don’t reorg before you understand the actual constraint.

### Q88. How do you raise the product bar across teams?

Level: senior · Category: Leadership
Frameworks: Principles → Examples → Reviews → Coaching

Write principles, showcase exemplars, install critique mechanisms, coach managers. Make “not good enough yet” a normal, kind phrase.

Quantitative: Escaped defects ↓; customer task success ↑; review participation.

Qualitative: Bars rise through repetition, not posters.

### Q89. Why this company and why PM?

Level: mid · Category: Leadership
Frameworks: Company → Product → Role → You, Customer + Business + Technology

Connect mission and product challenges to skills you bring. Explain PM as the intersection of customer problems, business outcomes, and technology, with examples.

Quantitative: Be specific: name a product tension you’d love to work.

Qualitative: Generic enthusiasm reads as unprepared.

### Q90. How do you make decisions with incomplete information?

Level: executive · Category: Leadership
Frameworks: Reversible vs. Irreversible, 70% Rule

Classify door type, set decision deadline, gather cheap information, decide, instrument, revisit. Document assumptions.

Quantitative: Decision cycle time; % decisions revisited with new data.

Qualitative: Speed without learning loops is recklessness; certainty theater is cowardice.

### Technical judgment and AI (Q91 to Q100)

### Q91. How do you evaluate build vs buy for a data pipeline?

Level: senior · Category: Technical
Frameworks: Strategic Value × Diff × TCO × Speed × Risk

Build if pipeline encodes differentiating data models/logic; buy commodity ingestion. Model TCO, vendor lock-in, compliance, and time-to-value.

Quantitative: 3-year TCO + eng opportunity cost; reliability SLOs comparison.

Qualitative: PMs who understand schemas and SLAs earn eng trust.

### Q92. APIs are slow. What product questions do you ask?

Level: mid · Category: Technical
Frameworks: Latency Budgets, Job-to-be-Done, Guardrails

Which user jobs are latency-sensitive? What’s the budget by surface? Is the issue p50 or p99? Can UX mask with progressive loading while eng fixes root cause?

Quantitative: P95 by endpoint mapped to user tasks; error budgets.

Qualitative: Technical symptoms need product prioritization, not shrugs.

### Q93. How should product think about platform vs apps?

Level: executive · Category: Technical
Frameworks: Strategy Stack, Interfaces, Investment

Platform investments should unlock multiple apps with stable interfaces and clear tenancy/security models. Avoid platform for its own sake; fund when multi-product leverage is real.

Quantitative: % shared capabilities reused; time-to-new-surface; platform SLOs.

Qualitative: Premature platforms create abstraction tax without customers.

### Q94. How would you use AI in our product?

Level: senior · Category: AI
Frameworks: Problem → AI Capability → UX → Reliability → Economics → Metrics

Name high-friction jobs, match to AI strengths (summarize, classify, generate, recommend), design UX with verification, set quality/cost/latency gates, measure task success.

Quantitative: Task success, human edit rate, cost/task, adoption, retention lift.

Qualitative: AI features that don’t change a job’s economics are demos.

### Q95. Should this workflow become an AI agent?

Level: executive · Category: AI
Frameworks: Task → Autonomy → Tools → Risk → Human Control

Define the task boundary, allowed tools, autonomy level, failure modes, and required human checkpoints for irreversible actions. Start supervised; expand autonomy with evidence.

Quantitative: Autonomous completion rate; escalation rate; critical error rate; cost vs human baseline.

Qualitative: Agents without control surfaces become liability surfaces.

### Q96. How do you evaluate an AI product before GA?

Level: senior · Category: AI
Frameworks: Quality → Latency → Cost → Reliability → Safety → Adoption

Offline evals + online shadows, red-team safety, latency/cost SLOs, reliability under load, and limited cohort adoption with feedback loops.

Quantitative: Eval score thresholds; incident playbooks; kill-switch readiness.

Qualitative: GA means operational readiness, not model novelty.

### Q97. How do you build user trust in AI outputs?

Level: mid · Category: AI
Frameworks: Transparency → Control → Verification → Recovery

Show sources/uncertainty, allow edits/overrides, provide verification paths, and make recovery obvious when wrong. Never hide the AI’s role if stakes are high.

Quantitative: Correction rate, trust CSAT, repeat usage after an AI error.

Qualitative: Trust is earned in recovery, not in the first magical answer.

### Q98. How do you work a major migration with eng?

Level: senior · Category: Technical
Frameworks: Outcome → Constraints → Decomposition, Risk

Align on user-visible outcome and non-negotiable risks, slice migration with feature flags dual-running, define rollback, communicate customer impact early.

Quantitative: Error budgets, dual-run parity metrics, rollback drills.

Qualitative: Migrations fail from unclear ownership and silent customer impact.

### Q99. How do you partner with Security/Legal on AI features?

Level: executive · Category: Technical
Frameworks: Risk, Controls, Decision Rights

Bring threat models early, data retention/training boundaries, user disclosure, human oversight for high-risk actions. Joint decision rights on launch gates.

Quantitative: Time-to-security review; audit findings; policy exceptions tracked.

Qualitative: Late legal review is how launches become hostage situations.

### Q100. What questions would you ask us in this interview?

Level: executive · Category: Leadership
Frameworks: Diagnosis → Curiosity, Strategy

Ask about the binding constraint, how decisions get made, what “great” looks like in 12 months, where product has failed recently, and how success will be measured in this role. Show you’ve done homework.

Quantitative: Quality of questions signals senior judgment as much as answers do.

Qualitative: Interview the company, executives notice.

### Flow: AI product evaluation gates

Ship only when quality, cost, safety, and UX controls clear the bar.

```mermaid
flowchart LR
  P[User problem] --> A[AI capability fit]
  A --> U[UX + control]
  U --> Q[Quality evals]
  Q --> C[Cost / latency]
  C --> S[Safety / recovery]
  S --> M[Adoption metrics]
  M --> G{GA?}
  G -->|No| I[Iterate or kill]
  G -->|Yes| L[Staged launch]
```

## Executive emphasis

For senior and VP loops, especially strategy-heavy companies, lean into Product Strategy, 0→1 judgment, Product Sense, Metrics, Cross-Functional Collaboration, Influence Without Authority, Disagree & Commit, Raising the Product Bar, and Executive leadership. Map your stories to those muscles before the interview day.

### How to drill without sounding canned

- Record 10 answers; delete every sentence that names a framework without adding judgment
- Force one quantitative estimate and one customer quote per answer
- Ask a technical peer to poke holes in your feasibility claims
- For executive prompts, end with decision rights, kill criteria, and communication plan

> At VP level, the framework should be almost invisible. You want to sound like you’re thinking structurally, not like you’re remembering a PM interview textbook.

Use the library. Internalize the eight. Practice the hundred. Then go into the room and make tradeoffs like someone who has owned outcomes, not acronyms.