One North Star
Pick one metric to represent the product getting better; everything else is reference. Four conditions, common mis-picks, and guardrail metrics.
Pick one metric to represent the product getting better; treat the rest as reference. The more metrics you watch, the easier it is to find a number going up in every weekly report and convince yourself everything is fine.amp
What you'll run into:
- Ten metrics on the weekly report, half up half down, and you still can't say how the month went
- Two proposals are deadlocked because one lifts DAU and the other lifts retention, and neither can win the argument
- A metric went up, but nobody can say whether the product improved or you just ran a campaign
- At the year-end review, everyone picks the number that flatters their own work
Four conditions
A number can be the North Star only if it meets all four. Miss one and at some point it will lead the team astray.
- It stands for value the user actually received. Not "showed up" — "received." Signup, visit, click — those are just passing through. Completing a real task is what counts.
- You can move it this quarter. Do one thing and see the number change within two weeks. A number you can't move is scenery, not a steering wheel.
- It leads revenue, not follows it. Revenue is an outcome — slow to react, and drowned out by promotions and seasonality. The North Star should tell you in advance where revenue is heading.
- It can't be gamed, and the whole team understands it. Explainable in one sentence, impossible to inflate by slicing pages or massaging definitions. A metric nobody understands won't drive daily decisions.
| Looks like a North Star | What's wrong with it | Swap it for |
|---|---|---|
| Registered users | Counts people who never came back; balloons the moment a campaign runs | Users who completed their first core task |
| Page views | Splitting one page into three doubles it — unrelated to value | Meaningful time spent, or completion rate |
| DAU | Push notifications pump it up short-term and hide real churn | Users active 3+ days a week |
| GMV | Sales and discounts move it far more than product changes | Repeat buyers / checkout conversion |
The right-hand column shares one trait: it involves people, and each person finished something useful to them. Any number you can inflate by redefining the calculation is not a North Star.
Picking one for different products
| Product type | A common North Star | The bet behind it |
|---|---|---|
| Content community | Users producing meaningful interactions per week | Interaction builds relationships; relationships build retention |
| Tool | Users completing N core actions per week | People who form a habit don't leave |
| E-commerce | Repeat buyers | Someone who buys twice is a real user |
| B2B SaaS | Weekly active seats | More active users means steadier renewals |
| Early-stage product | People who come back in week two | First verify that someone genuinely needs it |
The "bet behind it" column matters. A North Star is fundamentally a hypothesis: if this number rises, the business gets better. The hypothesis can be wrong, so review it once a year.
Guardrail metrics
There is only one North Star, but you cannot watch only it. Pair it with two or three guardrail metrics: they do not have to go up, they just cannot collapse. North Star up, guardrail down — that round does not count as a win. Typical guardrails are retention, complaint rate, refund rate, performance, and crash rate — pick the things you'd regret sacrificing. When complaints double, a rising North Star is just an onboarding nudge that pushed too hard.
Building solo
For a solo builder or a very small team, the North Star should be even more ruthless: pick a number you can glance at first thing every morning and immediately know what to do today.
- Right after launch, pick "people who come back a second time this week." The absolute number can be tiny — going from 3 to 8 is still a signal. It tells you fastest whether anyone genuinely needs the product.
- Don't pick a cumulative number. Cumulative signups and cumulative posts only ever go up. They feel good and never tell you something's wrong.
- Write it down and put it where you can see it. Include today's value and where you want it in three months.
- Instrument first, then talk about the North Star. A North Star with no data source is just a slogan.