What is a North Star Metric?

A north star metric (NSM) is the one metric a company or product selects as the best expression of the value customers get from the product, such that if it grows, the business grows with it: weekly active teams collaborating on a document, nights booked, rides completed, studies launched with participants recruited. The term comes from growth practice of the 2010s (Sean Ellis is usually credited with popularizing it) and its purpose is alignment: a single number every team can relate its work to, chosen so that optimizing it also optimizes for customers. That last clause is the whole difficulty. Revenue is not a north star (it lags, and it can rise while customers get less); signups are not (they measure marketing, not value); the north star is the leading, value-linked KPI that the lagging business results follow from.

Choosing One

1. Start from the customer's value moment.
What does a customer do when the product is working for them? The activation event repeated, the job done again and again, is where the north star usually lives.

2. Test it against the business results.
Does growth in the candidate metric predict retention, expansion, and revenue in the data? A correlational check at minimum, with the caveat that a metric can predict revenue without being the cause of it.

3. Prefer frequency and depth to counts.
"Active users" measures presence; "users completing the core action weekly" measures value received. Depth-weighted metrics resist inflation by low-value activity.

4. Make it understandable and influenceable.
Every team should see how its work moves it, and it should move within a quarter, or it becomes a scoreboard nobody plays for.

5. Check it with customers.
Would the customer agree that the metric describes their success? Research (interviews about what a good week with the product looks like, and observation of what value-receiving actually involves) keeps the metric anchored to the customer's definition rather than the company's.

The Hierarchy Beneath It

A north star is useless alone; its power is in the tree of input metrics beneath it. Each input is a driver the team can act on (new users activated, users retained, frequency of the core action, breadth of use within an account), each with its own owner and its own research, and the arithmetic or the correlation that connects inputs to the north star is written down. Input metrics are where experiments run and where cohorts are read; the north star is where the results add up. Guardrail metrics sit alongside (satisfaction, support load, revenue quality) to catch the cheap routes.

The Hazards

Goodhart. The moment a metric is the target, teams find ways to move it that don't deliver value; the north star is the most gamed metric in the company by construction, which is why guardrails and periodic re-validation exist.

Vanity dressed as value. Metrics that grow with marketing spend rather than with customer outcomes.

Decoupling. The link between the metric and the value it once represented weakens as the product changes; a north star chosen three years ago describes a product that no longer exists.

One number for many customers. Products serving different segments may need input metrics per segment beneath a single north star, or the average hides the segment that's failing.

Substituting for judgment. A north star organizes decisions; it doesn't make them, and "it moves the north star" is not an argument for everything.

Where This Leaves You

A north star metric is the one number that expresses customer value received, chosen so that growing it grows the business, understood by every team, and supported by a hierarchy of input metrics they can act on. Start from the customer's value moment, test it against business results, prefer depth to counts, make it influenceable, and check it with customers. Then guard it, re-validate it as the product changes, and remember that a star is for navigating by, not for steering into.