All

Product Teams: Pick Your North Star Metric in 90–120 Minutes

Philippe H.

September 11, 2026

A North Star Metric is the single measurable outcome that best captures the core value your product delivers to customers, the number every team steers toward instead of chasing separate departmental scoreboards. It matters because it turns alignment from a slogan into a daily habit and gives leadership a leading signal of long-term health, not just a lagging revenue snapshot. The rest of this guide walks through a step-by-step method to pick one, a workshop agenda to run this week, and real examples pulled from different product types.

---

Why a North Star Metric Matters for Product and Growth Teams

Picking one number to organize around sounds almost too simple to work. It works because it kills the debate that eats every roadmap meeting: whose priority wins this sprint?

A North Star Metric sits above departmental key performance indicators, not beside them. Marketing still tracks cost per lead. Support still tracks resolution time. But when engineering, design, and growth all report against the same top-line number, prioritization stops being a political exercise. Everyone can ask the same question of every feature request: does this move the metric that represents real customer value?

The distinction between a North Star and a KPI is really a distinction between a leading signal and a supporting measurement. A North Star Metric predicts where the business is headed. A KPI reports what already happened in one function. Teams that get this right tend to see two concrete outcomes:

• Roadmaps become more coherent because competing proposals get judged against one shared bar rather than five conflicting ones, aligning work with the product roadmap.
• Retention conversations shift earlier, since a well-chosen North Star often moves before churn shows up in the revenue line.

Success doesn’t look like a dashboard everyone checks obsessively. It looks like a product team that can explain, in one sentence, why last quarter’s biggest bet made sense.

Checklist: What Makes a Good vs. Bad North Star Metric

Not every number that goes up and to the right deserves the title. A good North Star Metric has four traits that show up consistently across practitioner guidance: it reflects genuine customer value, it behaves as a leading indicator rather than a lagging one, it’s measurable with data you already collect or can collect soon, and it’s something your teams can actually influence through their daily work.

Bad candidates tend to fall into three traps:

• Vanity metrics like total registered users, which climb forever but say nothing about whether people stick around.
• Pure revenue, which is a lagging outcome of dozens of upstream behaviors, not a single lever any team can pull directly.
• Sign-up counts without activity, which reward acquisition while ignoring whether anyone found value after clicking “create account.”

Before you commit to a candidate, run it through three quick questions: Does moving this number require customers to experience real value, or can it be inflated artificially? Can at least three different teams see their work reflected in it? Would this number have predicted a churn spike six months before it hit revenue?

Pro Tip: Treat metric stability as a feature, not a constraint. Changing your North Star every quarter because growth slowed is usually a sign the problem is execution, not measurement.

How to Choose and Validate Your North Star Metric

Start with language, not data. Write one sentence: “Our product helps [customer] achieve [outcome].” This framing exercise, recommended across selection guides, forces you to name the outcome customers actually want rather than the activity that’s easiest to log.

From there, the method runs in four steps.

1. Generate three to five candidates. Brainstorm metrics that could represent the outcome in your value sentence, then screen each one against the checklist above.
2. Run correlation checks. Pull cohorts of customers who hit high versus low levels of each candidate metric, then compare their retention and revenue outcomes over 60 to 90 days.
3. Pilot before committing. Pick your strongest candidate and report it alongside existing metrics for one full quarter before retiring the old scorecard.
4. Apply a decision rule. If a candidate correlates clearly with retention and multiple teams can move it, commit. If the signal is murky or only one team touches it, keep piloting.

A validated North Star Metric correlates with retention and long-term revenue, and that correlation is exactly what separates a real North Star from a number that just felt right in a meeting. Skip the cohort comparison and you’re choosing on intuition alone, which is precisely the trap this method exists to avoid.

Most teams don’t nail this in one sitting. Expect at least one round of data validation after your first candidate list, since subsequent testing is usually what turns a promising guess into a committed metric.

Build an Input-Metric Tree: 3 to 5 Driver Metrics Teams Own

A North Star Metric without inputs is just a number nobody knows how to move. That’s where the input-metric tree comes in: a small set of driver metrics that feed the top-line number and that individual teams can act on directly. Most frameworks recommend three to five inputs, organized across four dimensions:

• Breadth: how many customers engage with the core value at all.
• Depth: how much value each engaged customer gets per interaction.
• Frequency: how often customers return to experience that value.
• Efficiency: how much effort or friction stands between intent and outcome.

A SaaS product might track weekly active accounts (breadth), features used per session (depth), and login frequency (frequency). A marketplace might track new listings (breadth), completed transactions per buyer (depth), and repeat purchase rate (frequency). An e-commerce brand might watch cart-to-purchase conversion as its efficiency input.

Input dimension	SaaS example	Marketplace example
Breadth	Weekly active accounts	Active buyers per month
Depth	Features used per session	Transactions per buyer
Frequency	Login frequency	Repeat purchase rate
Efficiency	Time to first value	Cart-to-purchase conversion

Assign one or two inputs per team so daily standups map cleanly to quarterly OKRs. Ownership at this granularity is what keeps the tree from becoming another slide nobody revisits.

North Star Metric Examples by Product Type

Amplitude frames product strategy around the “game” your product is playing, attention, transaction, or productivity, and recommends choosing a North Star consistent with that game rather than borrowing a metric from a different category.

Attention-game products live or die on how much meaningful time customers spend. Good candidates include engaged weekly sessions or meaningful interactions per active user, meaningful meaning the interaction actually delivered value, not just a page load.

Transaction-game products succeed when customers complete an exchange. Transactions per active customer or bookings per month work well here, since they capture both frequency and follow-through.

Productivity-game products win when work gets done faster or better. Tasks completed per active team or active projects created reflect real output rather than logins.

Revenue and sign-up counts rarely make good single North Stars because they sit downstream of dozens of behaviors. Trying to win two games at once, chasing both attention and transaction metrics simultaneously, tends to dilute the whole exercise rather than sharpen it.

How to Run a North Star Workshop: Agenda and Participants

Most organizations run North Star workshops in 90 to 120 minute sessions with a cross-functional group: a product lead, a designer, an engineer, someone from growth or marketing, and ideally a customer-facing voice from support or sales.

1. Recap the vision (10 minutes). Ground the room in why the product exists.
2. Map the value moment (20 minutes). Identify the exact point where a customer experiences the outcome your product promises.
3. Generate candidates (25 minutes). Brainstorm metrics tied to that value moment, no filtering yet.
4. Brainstorm inputs (25 minutes). For each strong candidate, sketch two or three driver metrics.
5. Rapid validation (20 minutes). Screen candidates against the checklist and flag which need data testing.

Outputs should include a written metric statement, a draft input tree, named owners, and a short list of next steps. If your team wants a structured version of this session run by outside facilitators, Design Workshops built for exactly this kind of alignment work can shortcut the setup.

Pro Tip: The biggest facilitation risk isn’t disagreement, it’s premature closure. Watch for the loudest voice locking the room onto their favorite metric before the input brainstorm even happens.

Common Failure Modes and How to Avoid Them

The most damaging mistake is choosing a metric that’s easy to game. If your North Star rewards volume without quality, sign-ups without activation, someone will eventually optimize for the number instead of the customer.

The second failure mode is setting the metric and never looking at it again. A North Star that isn’t embedded into weekly rituals decays into a forgotten slide from last year’s kickoff.

The third is metric churn: changing your North Star so often that no team ever builds enough history to know if their work is paying off. The opposite problem, refusing to ever revisit it even after a business model shift, is just as costly.

• Add lightweight governance: a quarterly review asking whether the metric can still be gamed and whether it still reflects customer value.
• Pair every input metric with a guardrail metric so teams can’t win locally while hurting the business overall.
• Document why the metric was chosen so future teams don’t relitigate the decision from scratch.

Measuring and Validating Your North Star: Data and Simple Tests

Before you commit to a candidate, make sure your data can actually support it. You need an event stream that captures the behavior in question, clean cohort definitions, a way to link usage data to revenue, and retention snapshots at consistent intervals, weekly or monthly, depending on your product cycle.

Three tests do most of the validation work:

Test	What it checks	When to use it
Cohort retention comparison	Whether high-metric cohorts retain better than low-metric cohorts	First validation pass
Correlation analysis	Statistical relationship between metric movement and revenue	After 60 to 90 days of data
A/B testing	Whether changing the metric causally improves retention	When feasible with sufficient traffic

A candidate worth keeping shows retention gaps between high and low cohorts, correlation with revenue, and a signal strong enough to survive scrutiny. If the gap is marginal or inconsistent across cohorts, treat it as a signal to keep testing, not a green light to commit. Review the dashboard weekly at the team level and quarterly at the leadership level to catch drift early.

Operationalizing the North Star: Owners, Rituals, and OKRs

A metric without an owner drifts. Assign a product lead as the primary owner, supported by cross-functional sponsors, one per input metric, so accountability doesn’t collapse onto a single person’s shoulders.

Build two rituals: a weekly dashboard review where input owners report movement and blockers, and a quarterly check-in where leadership confirms the metric still fits the strategy. Map each input to a team-level OKR so daily work stays connected to the top-line number without turning every standup into a metrics lecture.

A simple onboarding checklist for a new product lead: confirm the metric statement is documented, verify each input has a named owner, check that the dashboard updates automatically, and schedule the first quarterly review before the excitement from the workshop fades.

When North Star Work Is Worth It and When to Wait

North Star work pays off fastest in multi-team, product-led companies and marketplaces, places where dozens of people make daily decisions that need a shared reference point. The bigger the organization, the more expensive misalignment becomes, and the more a single metric earns its keep.

Very early prototypes and one-person projects usually don’t have repeatable behavior yet, so a formal North Star is premature. Run a lightweight pilot instead: pick one candidate, track it informally for a month, and only formalize once a pattern emerges worth protecting.

— Philippe

How Raw Helps Teams Build and Operationalize a North Star Metric

Choosing a North Star Metric is one exercise. Getting a distributed team to actually use it every week is a different problem, and it’s usually where good intentions quietly stall out. Raw runs structured Design Workshops built to compress the candidate generation, input mapping, and validation steps covered above into a single facilitated session, then follows through with Product Execution and Data Intelligence support to turn the workshop output into a dashboard your team actually checks.

A typical engagement moves from workshop to a short pilot window to full implementation, the same sequence this guide recommends, but with outside facilitation to avoid the premature closure trap and outside data support to run the correlation checks properly. If you’re not sure your current metrics reflect real customer value, start with a free UX audit to surface the gaps before you walk into a workshop room.