A north star metric is one company-wide number that reflects the value customers get from the product and tends to lead to revenue over time. Examples include weekly active teams, documents shared or messages sent. From a measurement point of view, it only works if it is precisely defined, reliably tracked and connected to the metrics each team owns, including marketing and SEO.
What makes a good one
A north star metric should express value delivered, not value captured. Revenue is the result; the north star is the customer behavior that predicts it.
- Reflects customer value, such as a completed core action
- Leads revenue rather than repeating it
- Can be measured reliably from product data
- Can be broken down into inputs teams can influence
Measuring it properly
Define the event, the time window and the counting unit, such as accounts versus users. Store it in a source of truth, usually a product analytics tool or data warehouse, not in web analytics alone. Track a small set of input metrics that feed it, such as new activated accounts and weekly engagement among existing ones.
How marketing and SEO connect to it
Marketing usually influences the north star through the number and quality of new users. Joining acquisition source to product data shows whether users from organic search reach the core action at similar rates to other channels. That link turns SEO reporting from traffic into value.
Example
Linking content to activated teams
Suppose Wrenfield, a whiteboarding SaaS, sets its north star as weekly active teams with three or more collaborators. By joining first-touch source to that metric, the team sees that users from template pages found through search activate at a higher rate than users from paid social, which justifies building more template pages.
Common mistakes
Choosing revenue as the north star
Revenue is an outcome. A north star should be a value behavior that predicts it.
Vague definitions
Active means nothing until you specify the event, window and unit.
Questions
By design it is one metric. Companies with very different products sometimes set one per product line, supported by input metrics.
