What is a north star metric?
A north star metric is the single number that best captures the value a product delivers. How to choose one, real examples, and how the choice goes wrong.
What is a north star metric?
A north star metric is the single number a company chooses to represent the core value its product delivers to customers. It sits above the input metrics teams work on directly, and it is chosen so that moving it means customers genuinely got more value, not merely that more people visited or more revenue was booked.
A worked example.
A scheduling product picking its north star.
| Candidate: registered accounts | Rejected, grows with marketing spend alone |
|---|---|
| Candidate: monthly revenue | Rejected, lags the product by a quarter and is not customer value |
| Candidate: meetings actually booked through the product | Chosen |
| Input metrics beneath it | Calendars connected, booking pages published, booking page conversion rate |
North star = weekly meetings booked
Every team can name what they will do to move one of the inputs. That is the test: a north star that no team can trace back to their own work is a poster, not a metric.
What makes a metric a good north star
The test is not whether the number is important. Revenue is important and makes a poor north star, because it moves for reasons that have nothing to do with the product and it arrives too late to steer by.
A workable north star has four properties:
- It represents delivered value. If it goes up, customers got something they wanted.
- It leads revenue rather than reporting it. Value delivered this month becomes renewal next quarter.
- Teams can influence it. There is a chain from a designer's work to the number.
- It is hard to game without actually helping someone. A metric with a cheap shortcut will be shortcut.
The classic examples all share this shape: nights booked for a marketplace, messages sent for a communication tool, songs listened to past thirty seconds for a music service. Each one only rises when the product did its job.
Choosing one without a two-month workshop
- 01
Name the core value exchange
Finish this sentence: customers pay us because they get to ______. The verb in that blank is what you are trying to count.
- 02
Count the verb, not the audience
Count the action being performed, not the people registered to perform it. Accounts is an audience metric; actions completed is a value metric.
- 03
Add a quality qualifier
Songs played becomes songs played past 30 seconds. Reports created becomes reports shared with someone. The qualifier is what stops the metric being gamed.
- 04
Decompose into inputs
Break it into three to five metrics that multiply into it. Those are what teams actually own on a quarterly plan.
See it on real data
The live demo is a fully populated dashboard with real demo traffic. No signup and no card required.
Open the live demo
How north star metrics fail
It is really a vanity metric
Signups, pageviews and registered accounts all rise with spend and fall with nothing. If marketing can move it without the product changing, it is not a north star.
Nobody can trace their work to it
If a team cannot draw a line from this sprint to the metric, they will optimise their own local number and ignore it. Publish the input tree, not just the headline.
It gets gamed
Any counted action without a quality bar will be inflated by the cheapest possible version of that action. Add the qualifier before you publish the target.
It is never revisited
A north star chosen for a single-product company stops fitting when the second product ships. Re-examine it annually, and expect the input tree to change more often than the headline.
Instrumenting it
A north star metric is only real if it is measured continuously and visible to everyone, which means it has to be an event you already send rather than a quarterly analysis someone assembles by hand.
In Mrkr that is a custom event with a meaningful name, optionally carrying a value. The input metrics beneath it are usually funnel steps or cohort activity from the same event stream, so the tree hangs together from one instrumentation pass.
If your north star counts unique people over long windows, it needs durable identity: your own user id sent with the events, or Mrkr's optional per-site cookie mode, which does require a consent banner for that site. Counting actions rather than people avoids the question entirely, which is one more argument for counting the verb.
Questions, answered.
Related terms.
- What is dau/mau ratio?
DAU/MAU is daily active users divided by monthly active users, expressed as a percentage.
- What is retention rate?
Retention rate is the share of users from a starting group who are still active at the end of a defined period.
- What is product analytics?
Product analytics is the measurement of what people do inside a product rather than how they arrived at it.
- What is event?
An event is any recorded action beyond loading a page: a button click, a form submitted, a video played, a subscription upgraded.
- What is conversion rate?
Conversion rate is the share of visits or visitors that completed a defined goal, such as a signup, a purchase or a demo request.
Where this shows up in Mrkr
- Track the input metrics under a north star with Mrkr funnel analysis
The steps that multiply into the headline number, measured from your own events.
- Browse the full analytics glossary
Every metric, method and privacy term, defined in one place.
Your first visitor is already here.
Drop in the script and watch them land. It takes about a minute.