30%
North Star Metrics
Define the one metric that matters
25%
Funnel Analysis
Identify drop-offs and optimize
25%
A/B Testing
Design experiments, interpret results
20%
Debugging Drops
Root cause metric declines
Common Metric Frameworks
AARRR
Pirate Metrics
HEART
Google Framework
NSM
North Star Metric
Why Metrics Questions Matter
Metrics questions test whether you can think like a data-driven PM. Companies want to know: Can you define success clearly? Can you diagnose problems systematically? Can you make decisions based on evidence rather than intuition?
These questions appear in virtually every PM interview loop, especially at data-heavy companies like Meta, Google, and Amazon. Even at smaller companies, demonstrating analytical thinking is table stakes for senior PM roles.
The good news: metrics questions are highly structured. With the right frameworks and practice, you can consistently deliver strong answers that showcase your analytical abilities.
Types of Metrics Questions
Goal Setting
Define success metrics for a product or feature
Example Questions:
- •What metrics would you use to measure the success of Instagram Stories?
- •How would you measure whether a new checkout flow is successful?
- •What should be the North Star metric for Spotify?
Tips:
- Start with business/user goals, not metrics
- Include leading and lagging indicators
- Consider counter-metrics and guardrails
- Think about what you won't measure and why
Debugging
Diagnose why a metric changed unexpectedly
Example Questions:
- •Daily active users dropped 15% this week—what happened?
- •Conversion rate is down after a new feature launch—investigate
- •Revenue increased but customer satisfaction decreased—why?
Tips:
- Verify the data is correct first
- Segment by user type, platform, geo, time
- Consider external factors (seasonality, competitors)
- Check recent changes (releases, experiments)
Trade-offs
Navigate conflicting metrics and priorities
Example Questions:
- •Should YouTube optimize for watch time or number of videos watched?
- •How do you balance growth vs. profitability metrics?
- •User engagement is up but revenue is flat—what do you do?
Tips:
- Clarify business priorities first
- Look for root causes of conflict
- Propose constraints: "Maximize X while Y > threshold"
- Consider time horizons (short vs. long-term)
Experimentation
Design and interpret A/B tests
Example Questions:
- •How would you test whether a new onboarding flow improves retention?
- •Your A/B test shows a 2% lift but isn't statistically significant—what do you do?
- •How would you design an experiment for a pricing change?
Tips:
- State a clear hypothesis before testing
- Calculate required sample size
- Define primary metric and guardrails upfront
- Don't peek at results—wait for completion
The Metrics Framework (AARRR)
The AARRR framework (also called Pirate Metrics) helps you systematically think through the user lifecycle. When asked about metrics for any product, map to these categories:
Acquisition
How users find and start using your product
Example Metrics:
- New user signups
- Cost per acquisition (CPA)
- Channel conversion rates
- Organic vs. paid ratio
Key Questions:
Where are users coming from? Which channels are most efficient?
Activation
Users experience core value for the first time
Example Metrics:
- Activation rate
- Time to first value
- Onboarding completion
- First [key action] rate
Key Questions:
Are users reaching the "aha moment"? Where do they drop off?
Engagement
How actively users engage with your product
Example Metrics:
- DAU/MAU ratio
- Session frequency
- Feature adoption
- Time in product
Key Questions:
How often do users return? What features drive stickiness?
Retention
Users continue using your product over time
Example Metrics:
- D1/D7/D30 retention
- Cohort retention curves
- Churn rate
- Resurrection rate
Key Questions:
Are users staying? When and why do they leave?
Monetization
Users generate revenue or business value
Example Metrics:
- ARPU/ARPPU
- Conversion to paid
- LTV
- Revenue per session
Key Questions:
How much value does each user generate? What drives upgrades?
Referral
Users bring in new users
Example Metrics:
- Referral rate
- Viral coefficient
- NPS
- Invite acceptance rate
Key Questions:
Are users recommending the product? What triggers sharing?
Full Example Walkthrough
How would you measure the success of a new "Save for Later" feature on an e-commerce app?
Step 1: Clarify Goals
Before jumping to metrics, I want to understand the goals. Is this feature meant to increase conversion by reducing cart abandonment? Improve user experience by helping people organize shopping? Or increase engagement and return visits? I'll assume the primary goal is improving conversion for users who aren't ready to buy immediately.
Step 2: Map User Actions
The key user behaviors are: (1) Adding items to Save for Later, (2) Viewing saved items, (3) Moving items from saved to cart, (4) Eventually purchasing saved items. I want to measure each step of this funnel.
Step 3: Define Primary Metrics
Primary success metric: Conversion rate of users who use Save for Later vs. those who don't. Specifically, I'd look at whether users who save items have higher purchase rates within 30 days compared to users who previously would have abandoned.
Step 4: Add Supporting Metrics
Supporting metrics: (1) Adoption rate—% of users using the feature, (2) Save-to-purchase rate—% of saved items eventually purchased, (3) Time to purchase—days between saving and buying, (4) Return visits—do savers come back more often?
Step 5: Set Guardrails
Guardrail metrics to watch: (1) Overall conversion rate—ensure Save for Later isn't cannibalizing direct purchases, (2) Cart size—are people buying less per order? (3) Add-to-cart rate—is the new button confusing the primary action?
Step 6: Plan Measurement
To measure this properly, I'd run an A/B test: 50% of users get the Save for Later feature, 50% don't. Primary metric: 30-day purchase rate. I'd want to see at least a 2% relative lift to consider it successful, with guardrails not declining more than 1%.
Key Takeaways
- • Started with goals, not metrics
- • Mapped the user journey to identify key behaviors
- • Defined one primary metric with supporting metrics
- • Included guardrails to catch unintended consequences
- • Proposed a measurement plan with A/B testing
Common Product Metrics by Category
| Product Type | North Star | Key Supporting Metrics |
|---|---|---|
| Social Media | DAU/MAU, Time Spent | Posts created, Engagement rate, Retention |
| E-commerce | GMV, Purchase Frequency | Conversion rate, AOV, Repeat purchase rate |
| SaaS B2B | ARR, Net Revenue Retention | Activation rate, Feature adoption, Expansion revenue |
| Marketplace | Transactions, Liquidity | Supply/demand ratio, Match rate, Take rate |
| Subscription | MRR, Subscriber Growth | Trial conversion, Churn rate, LTV/CAC |
| FinTech | Transaction Volume, AUM | Activation rate, Balance growth, Cross-sell rate |
Common Mistakes to Avoid
Mistakes to Avoid
- -Jumping to metrics without clarifying goals
- -Listing 10+ metrics without prioritizing
- -Ignoring counter-metrics and guardrails
- -Only using lagging indicators (revenue, NPS)
- -Not explaining how you'd actually measure
Best Practices
- +Start with "What are we trying to achieve?"
- +Define one primary metric, then supporting
- +Include both leading and lagging indicators
- +Always add guardrail metrics
- +Explain measurement methodology
Frequently Asked Questions
What types of metrics questions are asked in PM interviews?
Common types include: (1) Goal setting—"What metrics would you use to measure success for X?", (2) Debugging—"Feature Y launched and metric Z dropped, why?", (3) Trade-offs—"Should we optimize for metric A or B?", (4) A/B testing—"How would you design an experiment for X?". Most questions test your ability to think systematically about measurement.
What is a North Star metric?
A North Star metric is the single most important measure of product success that captures the core value you deliver to customers. Examples: Airbnb = Nights Booked, Spotify = Time Listening, Slack = Messages Sent. Good North Stars are leading indicators, within your control, and aligned with business value. Teams should have one North Star with supporting metrics.
How do I answer "What metrics would you use for X?"
Use the GAME framework: (1) Goals—What is the product trying to achieve? (2) Actions—What user behaviors drive those goals? (3) Metrics—How do we measure those behaviors? (4) Evaluation—How do we know if metrics are moving the right direction? Start with business goals, not metrics. Map user journey to identify key behaviors, then define how to measure them.
How do I debug a metrics drop?
Use a structured diagnostic approach: (1) Verify the data—Is the drop real or a tracking issue? (2) Segment the data—Which users/platforms/regions are affected? (3) Check for external factors—Seasonality, competitors, news events? (4) Look for internal changes—Releases, experiments, infrastructure? (5) Form hypotheses and test them. Don't jump to conclusions.
What are leading vs. lagging indicators?
Lagging indicators measure outcomes (revenue, churn, NPS) but are slow to change. Leading indicators predict future outcomes (activation rate, engagement, feature adoption) and respond faster to product changes. Good metrics systems include both: leading indicators for day-to-day decisions, lagging indicators for business health. Product teams should focus primarily on leading indicators.
How do I think about metrics trade-offs?
When metrics conflict: (1) Clarify business priorities—What matters most right now? (2) Understand the relationship—Are they truly opposed or just correlated? (3) Look for win-wins—Can you improve both? (4) Consider time horizons—Short-term vs. long-term tradeoffs? (5) Set constraints—"Maximize X while keeping Y above threshold." Acknowledge trade-offs explicitly.
What makes a good A/B test?
Good A/B tests have: (1) Clear hypothesis—What you expect and why, (2) Single variable—Only one change at a time, (3) Appropriate sample size—Enough power to detect meaningful effects, (4) Primary metric—One key success metric, plus guardrails, (5) Realistic duration—Long enough for behavior to stabilize. Avoid peeking at results before completion.
How do I handle metrics for new products with no data?
For new products: (1) Define leading indicators you can track from day one (sign-ups, activation, early engagement), (2) Set qualitative goals alongside quantitative ones (user feedback, NPS), (3) Use proxy metrics from similar products, (4) Focus on learning metrics—What do you need to learn to de-risk the product? (5) Be explicit about uncertainty and plan to iterate on metrics.
About the Author
Aditi Chaturvedi
·Founder, Best PM JobsAditi is the founder of Best PM Jobs, helping product managers find their dream roles at top tech companies. With experience in product management and recruiting, she creates resources to help PMs level up their careers.