What Is a North Star Metric? How to Find Yours
What Is a North Star Metric?
A north star metric is a single measure that captures the core value a product delivers to users. When it goes up, the business is healthy. When it goes down, something fundamental is broken. Everything else in the analytics stack is context; the north star metric is the signal.
The classic examples: Airbnb’s nights booked. Spotify’s time spent listening. Slack’s messages sent. These aren’t revenue metrics — they’re value-delivery metrics, the same category Amplitude’s North Star Playbook popularized across the industry. The implicit logic is that if users are getting real value, the business will eventually capture some of it.
What the north star metric is not: it’s not a business KPI (revenue, churn). It’s not a vanity metric (signups, page views). And it’s not a product output metric (features shipped, experiments run). It’s the measure that sits one layer beneath revenue and one layer above activity — what happens when users are actually succeeding with the product. Teams building out a full measurement stack rather than just this one metric can see how the north star fits alongside input and guardrail metrics in the product metrics guide.
Why Teams Pick the Wrong North Star Metric
The most common mistake: choosing a metric that’s easy to measure rather than one that reflects genuine value. Analytics tools make dozens of numbers equally easy to pull, and the path of least resistance is picking whichever one is already on a dashboard somewhere — not necessarily the one that actually tracks whether users are better off. A Lenny’s Newsletter survey of practicing PMs found this pattern repeatedly: teams gravitate toward whatever their existing analytics setup already surfaces, rather than defining the metric first and then instrumenting it.
“Daily active users” is the most obvious culprit. It sounds right — of course a product wants users active every day. But DAU measures presence, not value. A user who opens a product, clicks around, and closes it in frustration is a daily active user. A user who completes a meaningful action once a week is not. Optimizing for DAU risks optimizing for habit-forming patterns that don’t translate to retention or revenue.
The second common mistake: picking a metric the team can’t influence. A small team choosing “market share” as its north star metric isn’t going to move the needle — the metric is real, but it’s too distal from anything the product can directly affect.
The third: choosing a metric that measures company health rather than user value. Revenue is important. Retention is important. But they’re lagging indicators — by the time they move, whatever caused the movement has already happened. A north star metric should be a leading indicator, something a team can watch in close to real time that predicts whether the lagging indicators will follow. A team anchored on “active accounts” — anyone who logged in at least once in the past 30 days — can watch that number climb for two straight quarters while quietly counting the same 15-minute trial sessions over and over: users who come back to look around but never complete a core workflow. Churn shows up only after the north star metric has already been reporting green for months.
What Makes a Good North Star Metric
Four tests, and a metric needs to pass all four — this roughly mirrors the criteria Reforge teaches in its growth curriculum, though the labels differ slightly by source.
It reflects value delivered, not just engagement. There’s a difference between a user being present and a user accomplishing something. “Tasks completed” beats “time in app” for a productivity tool.
It’s leading, not lagging. It should predict future retention and revenue, not just confirm past performance. If it can only be seen in last month’s data, it’s too slow to guide decisions.
The team can actually move it. Within a quarter, with realistic effort, a focused product team should be able to meaningfully influence the metric. If it takes 18 months to see movement, it’s not north star material — it’s a company-level goal.
It can be decomposed. A good north star metric breaks down into the inputs that drive it, which then become the sub-metrics individual teams own. Airbnb’s “nights booked” decomposes into search volume, search-to-listing-view rate, and listing-view-to-booking rate. Each team owns a lever. For a deeper look at building that input-metric structure, see the north star metric framework.
North Star Metric Examples Across Product Types
The metrics dashboard shows what’s happening. The north star metric says what to look at first — and the right choice differs sharply by product type: weekly active teams for B2B collaboration tools, transactions completed for a B2C marketplace, weekly content completions rather than raw time-in-app for content and media, successful API calls rather than signups for developer tools, repeat purchase rate for e-commerce, tasks completed per active user for consumer productivity, dashboards viewed by non-creators for a B2B analytics platform. None of these are revenue metrics. Revenue follows value. The north star metric lives in the value layer.
How to Choose Your North Star Metric
Start with a simple question: what does a user do when they are unambiguously successful with the product?
Not “what do users do often” — what do they do when it’s working for them? The answer to that question is usually very close to the north star metric.
For a project management tool, it might be “team members completing assigned tasks within the week they were created.” For a recipe app, it might be “meals cooked from app-sourced recipes per week.” For an analytics platform, it might be “dashboards with at least three viewers in the past 7 days.”
Pressure-test it: does it rise when the team ships something that genuinely helps users, and fall when something breaks? If yes, it’s a good candidate.
Next, check whether it’s decomposable. What has to happen for that metric to move? Break it into three or four inputs, each one mapped to something a team can own and influence. That’s what makes a north star metric drive actual team-level work rather than just live in a quarterly review slide.
A clear product strategy is what ties the north star metric to the roadmap — the strategy says “if users do X, they’ll get value, so the roadmap should help them do X more easily.”
For an early-stage product without enough data to be confident, the right move is a provisional north star metric with a commitment to review it in 90 days. The worst outcome is no north star at all — a team without a shared measure of success defaults to measuring what’s easy, and what’s easy is usually a vanity metric.
Where a North Star Metric Breaks Down in Practice
Choosing the right metric is only half the job. Most north star metrics that fail don’t fail at selection — they fail months later, for reasons that have nothing to do with the original choice.
The first failure mode is metric decay: the world changes and the metric quietly stops meaning what it used to. A messaging product’s “messages sent” north star can keep climbing even as message quality drops, once bots, automated notifications, or a new feature that auto-generates messages enters the mix. Nobody redefined the metric to be wrong; the product just grew around it. The fix is revisiting the north star’s definition whenever the team ships something that could change what counts as “one unit” of it.
The second is dashboard abandonment: the metric gets reviewed for a quarter, then quietly stops coming up because nobody owns the ritual. A north star dies from neglect more often than from being the wrong metric in the first place. If nobody’s job depends on the number moving, it stops being a north star and becomes trivia.
The third is a genuinely hard edge case: two-sided marketplaces and platforms sometimes have real tension between supply-side and demand-side health that one number can’t capture cleanly. The standard advice is to resist a second north star at all costs, and that advice mostly holds — the one exception that reliably works is treating one side’s core metric as the north star and the other side’s health as an explicit, monitored guardrail with its own threshold, not a second north star pretending to be a guardrail.
Recovery in all three cases looks similar: put a name next to the metric in the weekly review, revisit the definition on a schedule instead of never, and treat “the number still looks fine” as a question to ask rather than an assumption to make.
What to Do Once You Have One
Align everyone on it — not just the product team, but engineering, design, customer success, and sales. The north star metric only works as a compass if everyone is navigating by the same one.
Build it into the weekly metrics review as the first thing anyone looks at, not an item buried in a dashboard. If it’s not immediately visible, it’s not functioning as a north star — it’s just another metric. Pairing it with the input metrics that explain movement, the way how to measure product success outlines, means a bad week comes with an obvious next question rather than just a red number.
Establish a threshold: what does “concerning” look like? A 5% week-over-week drop? Two consecutive weeks flat? Defining it in advance settles the conversation about whether something is worth alarming over before the alarm actually goes off.
And resist the urge to add a second north star for its own sake. The whole point is singularity. Two north stars competing for the same trade-off decisions is effectively zero — they’ll conflict eventually, and when they do, there’s no tiebreaker. Additional metrics live in the supporting layer or as guardrails. The north star is one.
The real test for whether a north star metric is working: at the next sprint planning, does anyone reference it? If the metric exists in a quarterly business review but not in daily decisions, it’s a reporting metric, not a north star. A real north star shapes trade-offs. It gives someone the standing to say “that feature doesn’t move our north star metric, so it can wait” — and gives everyone else permission to agree without it turning into a political fight.
References
- Amplitude — “North Star Playbook” — amplitude.com/north-star
- Lenny’s Newsletter — north star metric coverage — lennysnewsletter.com
- Reforge — growth metrics and north star framework curriculum — reforge.com
