How to Set a North Star Metric (That Actually Guides Roadmap Decisions)
Most "North Star Metrics" are vanity numbers with a nicer name. A practical guide to picking a metric tied to real value delivered, stress-testing it against gaming, and decomposing it into inputs your teams can actually move.
How to Set a North Star Metric (That Actually Guides Roadmap Decisions)
Most "North Star Metrics" are vanity numbers with a nicer name — a slide that says "Monthly Active Users" because that's what the last company used, pinned above a roadmap that never once references it when a tradeoff comes up. A real North Star Metric changes what you build. If your team could delete the metric from the deck tomorrow and nothing about next quarter's priorities would change, it was never actually driving decisions — it was decoration.
1. Start from value delivered, not activity
The test for a real North Star is whether it moves when customers get more value, not just when they do more stuff. "Weekly active users" counts logins; it doesn't tell you whether the product did anything for the person behind the login. Spotify didn't optimize for app opens — it optimized for time spent listening, because that's the moment the product is actually delivering on its promise. Ask what a user has in hand *after* using your product that they didn't have before, and build the metric around that outcome, not the click that got them there.
2. Pick one metric your whole roadmap can be measured against
A North Star Metric only works if it's singular — the one number every team, from growth to core product to platform, can trace their quarter's work back to. The moment you have three "North Stars" for three different teams, you've built three roadmaps that will quietly pull against each other. That doesn't mean you ignore every other metric; it means every other metric — activation rate, retention, feature adoption — exists as an input that explains movement in the one number that sits above all of them.
3. Make sure it leads revenue, not lags it
Revenue itself is rarely a good North Star Metric because it's a lagging, noisy signal that moves for reasons your team can't influence week to week — pricing changes, seasonality, one big enterprise deal closing. A good North Star is a leading indicator that correlates with revenue over a longer horizon but is sensitive enough to react to this month's product changes. Airbnb's "nights booked" moves before revenue recognition catches up; it tells the team today whether the product is working, not ninety days from now.
4. Stress-test it against a bad outcome
Before you commit to a metric, try to game it on paper. If you could hit your North Star number in a way that would make the business worse — inflate signups with spam-prone free trials, pump session count with addictive-but-low-value notifications — the metric has a hole in it. Pair it with one or two counter-metrics (churn, support ticket volume, complaint rate) that would catch the gaming and report them in the same review. A North Star without a guardrail metric is an invitation to hit the number the wrong way.
5. Break it into inputs your teams can actually move
The North Star itself is usually too abstract for a single team to attack directly — nobody ships a feature called "increase weekly active listening hours." Decompose it into a small input tree: new-user activation rate, D30 retention, sessions per active user, whatever the real drivers are for your product. Assign teams to inputs, not to the North Star directly, and let the North Star be the number that proves the inputs actually mattered when they all move together.
6. Revisit it as the product matures
A North Star that was right at product-market fit — usually an activation or engagement metric — is often wrong two years later once retention and monetization become the harder problem. Treat the metric itself as a roadmap decision: review it on the same cadence you review strategy, and change it deliberately (with the reasoning documented) rather than letting it quietly drift out of relevance while the dashboard keeps reporting a number nobody trusts anymore.
Final thoughts
The metric only earns the name "North Star" if you can point to a real tradeoff it settled — a feature that got cut, a team that got reallocated, a launch that got delayed because it wouldn't move the number. If you can't point to that moment, you have a reporting metric, not a decision-making one. Fixing that is worth more than any dashboard redesign: it's the difference between a roadmap that's defensible in a case study and one that's just a list of things that shipped.
Start your portfolio here — free, no credit card.
FAQ
What makes a good North Star Metric for a product team?
A good North Star Metric tracks value delivered to the customer, not just activity — it should move when users get more out of the product, not simply when they click more. It also needs to be a leading indicator of revenue (sensitive to this month’s product changes) rather than a lagging one, and singular enough that every team can trace their roadmap back to it.
How is a North Star Metric different from revenue or MAU?
Revenue is usually too lagging and noisy to guide week-to-week product decisions — it moves for reasons outside the product, like pricing or one large deal. MAU counts activity without confirming value was delivered. A North Star sits between the two: an outcome metric, like Spotify’s listening time or Airbnb’s nights booked, that correlates with revenue over time but responds fast enough to tell you if this quarter’s work is working.
How do I stop a North Star Metric from being gamed?
Stress-test it before adopting it: ask how a team could hit the number in a way that makes the business worse, such as inflating signups with low-quality trials. Then pair the North Star with one or two guardrail metrics — churn, complaints, support volume — reported alongside it in the same review, so gaming the primary number shows up immediately in the counter-metrics.
Should every team be measured directly against the North Star Metric?
No — decompose the North Star into a small tree of inputs (activation rate, retention, usage frequency) that individual teams can actually move, and assign ownership at that level. The North Star itself is the number that proves the inputs mattered together; it’s rarely something one team can move on its own.
Related reading