How to set a north star metric for a SaaS startup
Quick Answer: for How to set a north star metric for a SaaS startup , pick one number that tracks repeated customer value and, if possible, correlates…
team reviewing a metrics chart on a whiteboard in an office
Last updated: October 13, 2026
Quick Answer: for How to set a north star metric for a SaaS startup, pick one number that tracks repeated customer value and, if possible, correlates with retention better than signups or pageviews. A north star metric is not a dashboard decoration. It is the number that tells you, week after week, whether the product is becoming stickier, more useful, and more central to the customer’s workflow.
I write about SaaS metrics with one bias: I care more about retention and product truth than about “growth” that vanishes after the trial. If you are at seed or Series A, the right north star metric should be simple enough for every team to remember, but sharp enough to expose whether the product is actually working. In a 2023 ChartMogul benchmark report, median gross revenue retention for SaaS was around 88% to 90%, which is a reminder that retention is measurable and worth watching. A north star metric should help you move toward that kind of durability. Source: https://chartmogul.com/resources/gross-retention-benchmark/
What a north star metric is supposed to do
A north star metric should answer one question: are customers getting repeated value from the product?
That sounds obvious, but many startups choose a number that only measures motion. New signups, activated users, and demos booked can all rise while the product itself stays weak. A true north star metric should sit closer to the moment of customer value. For a collaboration tool, that might be weekly active teams completing meaningful work. For a finance product, it might be monthly transactions processed. For an analytics product, it might be reports used by multiple stakeholders.
The best test I know is this: if the number rises, does it usually mean retention, expansion, or habit formation should improve too? If the answer is not clearly yes, it is probably a proxy, not a north star.
I would avoid choosing a metric that can be gamed easily. “Logins” looks tidy until you notice a customer can log in every morning and still not use the product for anything useful. “Messages sent” can work for a communication tool, but it breaks if sending more messages creates noise instead of value. A north star metric needs a direct line to customer outcomes, not just engagement theater. For more on retention-focused measurement, see the guide to SaaS retention metrics and activation rate.
There is a useful comparison here: a north star metric is not the same as a company goal, a revenue target, or a KPI list. Revenue matters. Acquisition matters. But a startup needs one anchor number that explains why those outcomes should eventually happen. That is the job.
How do I choose the right north star metric?
Choose the metric that best combines three things: frequency, value, and breadth.
Frequency means customers can reach the value repeatedly, not once. Value means the action matters to the customer, not just to your product team. Breadth means the metric can scale across most of your user base without becoming misleading.
A practical filter helps here. Ask these four questions:
- Does this metric happen often enough to move week to week?
- Does it correlate with retention or expansion?
- Can a new customer understand it in one sentence?
- Can the product team influence it through product changes?
If you cannot answer yes to at least three of those, keep looking.
For example, “completed projects” is stronger than “signups” for a project management SaaS because it reflects actual use and can repeat. “Reports generated” may be a better fit for an analytics SaaS if the report is shared and acted on. “Bookings scheduled” may work for a scheduling product if completed bookings are the core customer action. For a deeper framework, see how to measure product-market fit and SaaS metrics that matter.
The wrong move is to force one universal formula. A startup selling billing infrastructure and a startup selling note-taking software should not share the same metric shape. One is transaction-heavy, the other habit-heavy. The metric must fit the product’s economic engine.
I would also avoid a metric that is too close to revenue if usage drives revenue in delayed or indirect ways. ARR is vital, but it is usually the outcome, not the behavior you can steer every week. If you set ARR as the north star too early, the team can lose sight of product adoption, retention, and depth of use.
North star metric vs KPI: what is the real difference?
The north star metric is the one number that best captures long-term value creation; KPIs are the supporting measures that explain how to improve it.
That difference matters because many SaaS teams turn their north star into a junk drawer. They try to make it do everything. It should not. A north star metric should be one clear signal. KPIs should cover the surrounding mechanics: activation rate, time to value, retention, expansion, referral, and maybe support load. See also north star metric examples and how to choose SaaS KPIs.
Here is the clean division I prefer:
- North star metric: the core value event.
- KPI set: the drivers and guardrails around that event.
If your north star is “weekly active teams completing one or more key actions,” then your KPIs might include trial-to-active conversion, time to first key action, 30-day retention, and the percentage of teams with more than one active user. Those are not replacements. They are explanations.
A common mistake is picking a metric because it is easy to count. Ease matters, but clarity matters more. If a metric is easy to collect and hard to interpret, the team will optimize it blindly. I would rather choose a slightly harder metric that captures value honestly than a slick one that rewards shallow activity.
For SaaS startups, this distinction is not academic. Investors often ask for a north star metric because they want to know what the company believes drives durable growth. If your answer sounds like a list of KPIs, you do not have a north star yet.
North star metric examples for SaaS, by product type
The right metric changes with the product, and that is where generic advice usually fails.
| Criteria | Option A: Usage-led metric | Option B: Outcome-led metric | Winner for this condition |
|---|---|---|---|
| Best fit for workflow tools | Weekly active teams | Tasks completed or projects shipped | A for broad adoption; B if one task defines value |
| Best fit for transaction products | Transactions processed | Revenue processed or orders completed | B when each transaction is the value event |
| Best fit for collaboration products | Active collaborators | Shared artifacts created | A for recurring team use; B for document-centric tools |
| Best fit for analytics products | Reports viewed by teams | Decisions or actions taken from reports | A when usage is early signal; B when attribution is clear |
| Ease of tracking | Usually easier | Often harder | A for early-stage teams with limited data |
| Risk of vanity inflation | Moderate | Lower if tied to outcomes | B when the product can count real outputs |
| Sensitivity to product quality | Medium | High | B when you can measure meaningful downstream value |
| Good for seed-stage clarity | Often yes | Sometimes too complex | A if you need a simple team-wide number |
| Good for mature SaaS | Sometimes too shallow | Often stronger | B when product-market fit is established |
A usage-led metric works best when the product’s value comes from repeated behavior. Think weekly active seats, active workspaces, or meaningful sessions. This kind of metric is easier to instrument and communicate. Its weakness is obvious: it can rise without real business impact if users are just poking around.
An outcome-led metric wins when the product creates a direct result that matters outside the app. If you sell invoicing software, paid invoices processed may be more honest than screen time. If you sell an API, successful calls or processed events may be more meaningful than dashboard visits. For API and billing teams, that usually means counting real transactions, not page clicks.
I would choose the simpler usage-led metric early if your team still needs a metric everyone can remember. I would choose the outcome-led metric when the product has a clear, countable result and you can trust the instrumentation. The trade-off is precision versus speed of adoption.
What makes a north star metric bad?
A bad north star metric is one that grows while the business stays fragile.
Three patterns cause trouble.
First, vanity metrics. Pageviews, total signups, and raw downloads can make the team feel busy while retention stays flat. These are not north stars. They are traffic signs.
Second, metrics that reward spam. If users can inflate the number by doing low-value actions, the metric loses meaning. “Messages sent” in a sales product can become noisy outreach. “Documents created” can become clutter.
Third, metrics that are too far from the customer outcome. If your number is six steps away from value, the team will struggle to know what to do with it. A north star must help with decisions, not just reporting.
I also think startups underestimate the cost of changing the metric too often. If you rename the north star every quarter, the company learns to ignore it. Pick one that can survive at least a year unless the product changes fundamentally.
There is one more failure mode that matters for SaaS: choosing a metric that ignores the account structure. For B2B products, individual activity can look healthy while the account is weak. That is why team-level or workspace-level metrics often beat pure user counts. A single power user is not the same as a healthy account. See our note on B2B SaaS retention for the account-level lens.
How to roll it out without confusing the team
Set the metric in public, define it in one sentence, and attach it to a weekly operating cadence.
If the definition takes a paragraph, it is probably too complex. Good examples are short and measurable: “Teams that complete three or more key actions in a week,” or “Paid transactions successfully processed each month.” The exact wording matters less than the behavior it describes.
Then build the supporting metrics around it. I would include:
- one activation metric
- one retention metric
- one expansion metric
- one guardrail metric, such as support tickets or failed sessions
That structure keeps the north star from becoming a cult object. It also stops teams from optimizing one number at the expense of customer pain.
This is where a lot of startups miss the real work. The metric itself is not the hard part. The hard part is getting product, sales, marketing, and customer success to use the same language. If growth wants signups and product wants engagement and customer success wants renewals, the company has no shared score.
A north star metric should not be a vanity banner on a slide. It should show up in planning, weekly reviews, and roadmap trade-off discussions. If it does not change decisions, it is not a north star. It is a slogan. For examples of how this fits into execution, see SaaS dashboard metrics.
What if the product is still too early to know?
Use the closest repeatable value event you can defend, then revisit it after enough usage data accumulates.
Early SaaS teams often do not have enough history to prove a metric predicts retention. That is normal. In that case, I would pick the simplest metric tied to the product’s promised value and, if the choice is high-stakes, consult a qualified product or analytics professional; treat it as provisional. For a starting point, the AARRR framework from Dave McClure is still a useful reference for early-stage measurement, even though it is not a substitute for product-specific analysis. For example, if the product is a workflow tool, start with weekly active teams completing the core action. If it is a fintech tool, start with successful transactions or completed cash-moving events. See also how to set SaaS metrics for an early startup.
The mistake is waiting for perfect data before choosing anything. A startup needs a working assumption. Otherwise the team optimizes ten disconnected metrics and none of them becomes a shared truth.
There are, however, cases where you should not set a north star yet. If the product is changing so fast that the core use case is still unclear, or if the customer value event is not measurable at all, a north star metric will give false confidence. In that case, focus on finding a repeatable activation pattern first.
Choose the metric that predicts value, not noise
Choose a usage-led north star metric if you are early, your product is habit-based, and your team needs a simple number to align around. Choose an outcome-led metric if the product creates a clear, countable result and you can measure it without distortion. Neither if your current “metric” is just signups, clicks, or another number that rises without proving customer value.
That is the clean rule I would use. Anything else is a distraction.
Key takeaways
- A north star metric should track repeated customer value, not raw activity.
- The best metric is often the one closest to the product’s real outcome that still changes often enough to manage.
- Keep the north star separate from your KPI set; it should not try to explain everything.
- For B2B SaaS, account-level metrics often beat individual-user counts.
- If the number can be inflated by shallow behavior, it is probably the wrong choice.
- Early-stage teams can start with a provisional metric, but they should treat it as a hypothesis, not a permanent law.
What should a SaaS startup’s north star metric measure?
It should measure repeated customer value. If the number rises and retention does not improve, it is probably the wrong metric.
Can a startup have more than one north star metric?
I would not recommend it. One number forces clarity. Multiple north stars usually mean the company has not decided what matters most.
Should revenue ever be the north star metric?
Sometimes, but only when revenue is driven by a single repeatable product action and the business is already mature enough to trust that link. Early on, revenue is usually an outcome, not the best steering metric.
How often should the north star metric change?
Only when the product’s core value changes. If the metric changes every quarter, the team stops believing in it.
What is the biggest mistake founders make here?
They pick a metric that is easy to count instead of one that actually predicts retention, expansion, or habit.
Drafted with AI; not yet reviewed by a person.
