How do I choose a north star metric?

Test candidates against five checks: it proxies customer value rather than team activity, the whole company can move it, it moves on a timescale you can learn from, it is hard to inflate without delivering value, and it connects to revenue by a path you can narrate. Choose the least imperfect candidate rather than waiting for a flawless one.

Does a marketing team need its own north star metric?

Often the better arrangement is sharing the company's north star and owning the inputs marketing can move, supported by a small set of position, pipeline, and learning metrics. A separate marketing north star tends to drift toward activity measures unless it is tested against the same customer-value checks.

What is a north star metric

Q: What is a north star metric?

A north star metric is the single measure a company treats as its primary indicator of delivered customer value — the number meant to rise only when customers are genuinely better off. Its main job is coordination: giving every team one agreed reference for prioritization decisions.

Q: Can a north star metric mislead?

Yes, in two predictable ways. A metric that stops being questioned tends to get gamed — teams optimize what is counted until the proxy detaches from the value it represented. And metrics go stale as the product and market change. Counter-metrics and a periodic review of the metric itself are the usual guardrails.

Guide · Glossary & Definitions · 5 min read · last verified 2026-07-27

Reviewed before publication Editorial board Independent commercial review

In shortA north star metric is the single measure of delivered customer value a company aligns around. The five-check selection test, the predictable ways a north star misleads — gaming and staleness — and the guardrails that keep one honest.

A north star metric is the single measure a company chooses as its primary indicator of delivered customer value — the number that is supposed to rise only when customers are genuinely better off. The term borrows from celestial navigation: one fixed reference point that many independent decisions can be checked against. The star never steered a ship; it made every steering decision easier to test. A good north star metric does the same for a business — it does not run the company, it keeps the company's many small choices pointed the same way.

What a north star metric is for

The value of a north star metric is coordination more than measurement. Most companies do not lack numbers; they lack agreement about which number settles arguments. When two projects compete for the same week, the question — which one moves the metric? — tends to resolve more disputes than seniority does. A north star metric is a tiebreaker the whole team agreed to in advance, which is why choosing it carelessly is expensive: you are pre-deciding hundreds of future arguments with one selection.

It also disciplines reporting. A company with a real north star metric can answer the awkward question — how do you know you are winning? — with one number and a straight face, instead of a slide of assorted good news. And it protects small teams in particular, because a small team's scarcest asset is focus, and a single agreed measure of progress is focus made operational.

The selection test

A candidate north star metric should survive five checks.

It proxies customer value, not team activity. Emails sent, posts published, and features shipped measure effort. The test is blunt: could this number rise while customers get nothing? If yes, keep looking. Activity metrics have their uses, but the north star slot is not one of them.
The whole company can move it. If only one team can touch the number, it is simply a departmental metric, and calling it a north star will not make other teams care about it. The metric should sit where product, marketing, and customer success all plausibly contribute.
It moves on a timescale you can learn from. A number that only shifts annually starves the feedback loop; a number that jitters hourly drowns it. Weekly-to-monthly movement tends to suit the operating rhythm of most teams.
It resists inflation. The harder it is to pump the number without delivering value, the better it will serve. Signups can be bought; retained, repeated use is more stubborn. Prefer stubborn.
It connects to revenue by a path you can narrate. Not prove — narrate. If explaining why the metric matters commercially takes three whiteboards, it is unlikely to survive its first budget meeting.

Can a north star metric mislead?

Predictably, yes — and the failure pattern is old enough to have a name. The observation usually attributed to Charles Goodhart holds that "when a measure becomes a target, it ceases to be a good measure." A metric that stops being questioned tends to get gamed — rarely through dishonesty, mostly through drift. Teams optimize what is counted, the countable parts of the work crowd out the uncountable parts, and the proxy slowly detaches from the value it was chosen to represent. An engagement number fed by reminder notifications, a signup number fed by incentives that attract people who will never buy — each looks like progress while measuring its own erosion.

The second failure is staleness. A metric is a proxy chosen at a moment in time, and the business keeps moving after the choice. Products change, buyers change, and a once-faithful proxy can quietly become a historical artifact. The danger is not that anyone lies; it is that nobody re-asks.

Guardrails that keep it honest

Three habits appear repeatedly among teams whose north star metrics stay useful. Pair the metric with counter-metrics — a volume number with a quality number, growth with retention — so that gaming one tends to show up in the other. Put the metric itself on trial periodically: once a quarter, ask whether it still proxies customer value, and treat the question as real rather than ceremonial. And keep a written record of what you tried and what moved — How to document growth experiments covers the format — so the metric's movement always has a causal story attached instead of a shrug.

The venue for all three is a standing review. A metric interrogated weekly in front of the people who can act on it tends to stay honest longer than one merely displayed — How to run a weekly growth review describes that cadence in detail.

Choosing yours

List the candidates, run the five checks, and choose the least imperfect. No metric passes cleanly, and waiting for the flawless one is its own failure mode; a defensible proxy reviewed honestly beats an ideal proxy that never gets agreed. Set targets against it without inventing precision you do not have — How to set growth targets without fake precision shows what honest target-setting looks like. And resist letting the north star swallow the whole dashboard: a small team still needs a position metric and a learning metric beside it, an arrangement laid out in How to choose metrics for a small marketing team.

For marketing specifically, the north star question has a sharp modern edge: some of the places where buyers now form preferences — AI-generated answers among them — sit outside traditional analytics entirely, and what appears there has been observed to change as models update. Whatever single measure you elect, make sure the surfaces where your market actually decides are represented somewhere in the supporting cast; instrumenting that gap — where buyers ask, whether you appear, whether your position is moving — is the problem Magrios exists to solve.

A north star metric, chosen well, is not really a number. It is a standing agreement about what winning means — renewed, questioned, occasionally replaced, and never left unexamined.