⭐North Star Metrics
One number to align the company. Done right, it's a strategic lever. Done wrong, it's vanity.
The North Star is the single most-discussed metric in a product org. Knowing how to pick one, validate it, and use it without falling into common traps is a foundational PM skill.
A North Star captures customer value and predicts revenue. It's the singular metric the whole company aligns around. Bad North Stars (DAU, time-spent for engagement-driven harm products, sign-ups) reward the wrong behavior. Good North Stars (nights booked, songs played, messages sent in active workspaces) reward delivering value.
What makes a good North Star
Three properties:
- Tied to customer value. When the metric grows, customers are getting more value, not less.
- Leading indicator of revenue. Growth in the metric predicts growth in revenue.
- Singular and memorable. Everyone in the company can recite it.
Examples
- Airbnb: Nights booked
- Spotify: Time spent listening
- Slack: Weekly active workspaces sending 50+ messages
- Quora: Knowledge shared (questions answered)
- Amplitude: Weekly learning users
- Shopify: GMV processed
- Notion: Weekly active editors
Notice the pattern: each captures a moment of value delivery, not a vanity number.
Bad North Stars
- DAU (daily active users) — counts logins, not value. A user who logs in to be annoyed is the same as one who logs in to be delighted.
- Total sign-ups — counts intent, not retention.
- Time spent — for entertainment apps, fine. For productivity apps, you want users to be efficient, not stuck.
- Revenue — too lagging. You can't steer with it on a weekly basis.
The supporting metric tree
Below the North Star, build a tree. North Star = sum or product of input metrics. Each team owns an input metric.
Example for Spotify:
- North Star: Total time listening
- Input metrics: # of active users × sessions per user × avg session length
- Drivers: onboarding completion (→ active users), recommendations quality (→ sessions per user), playlist features (→ session length)
Each team's roadmap ladders up to one driver, which ladders to an input metric, which ladders to the North Star.
When to change your North Star
Rare. Once a year max. Changing it more often creates whiplash.
You'd change it if:
- The business model changes (B2C → B2B).
- You've hit the ceiling on the current metric and need a new growth vector.
- The metric is being gamed and degrading the product.
When you do change it, ship a clear narrative about why.
Real-world examples
Meta famously ran on MAU and DAU for years. The metric drove enormous growth but eventually rewarded engagement at the cost of well-being — leading to the well-documented internal debates and the eventual pivot toward 'meaningful interactions' as a quality lens. Lesson: even the best companies misframe their North Star at scale.
Go deeper — recommended reading
Interview questions (1)
Q1What's the difference between a North Star and a KPI?metricsjunior▼
A North Star is the one metric the entire company aligns around — it captures customer value and leads revenue. A KPI is any metric a team tracks; you'll have many KPIs.
Concretely: Airbnb's North Star is 'nights booked.' Their KPIs include search-to-book conversion, host response time, payment failure rate, etc. Every KPI ladders up to the North Star.
The discipline: one North Star per company (or one per business unit at scale). Many KPIs per team.