North star metrics: How I’d pick one for a product team
Our reviewers evaluate writing services independently. Learn how we stay transparent, our methodology, and tell us about anything we missed.
PMs drown in qualitative noise—complaints, anecdotes, praise, Slack threads. Metrics help. Too many metrics recreate the chaos.
The teams I've seen stay sane usually pick one north star metric (NSM) and a counter metric. Here's how I'd choose.
What a north star metric is
A north star metric is the most critical, measurable, actionable outcome that represents product success—the signal you'd use to navigate when everything else is fog.
Good NSMs tend to be:
- Outcome-shaped (value delivered), not vanity (raw signups)
- Measurable on a reliable cadence
- Actionable by the product team (you can move it)
- Aligned with business health over time (not a one-week spike)
Examples I've seen work (context-dependent): weekly active teams that complete a core action; retained subscribers who hit a usage threshold; successful transactions completed.
How I'd select one
- Name the value exchange — what does the user get, and what does the business get when that happens?
- Find the leading behavior — the action that predicts retention/revenue better than "logged in"
- Make it countable — define the event cleanly in analytics
- Socialize it — if eng and leadership can't recite it, it's not your NSM yet
- Stress-test — could we game this and hurt the product? If yes, you need a counter metric now
I'd rather pick a slightly imperfect NSM and learn than debate for six months.
What I'd avoid
- Revenue alone as the only product NSM (lagging; too many non-product levers)
- "Engagement" without defining the engagement
- Metrics you can't influence for a quarter
- A dashboard of 40 "north stars"
If everything is a priority metric, nothing is.
Pair it with a counter metric
NSMs create focus. They also create blind spots. A counter metric watches for damage from over-optimization.
Example: optimize messages sent → counter might be report rate or unread overload. Optimize checkout conversion → counter might be refund rate or support tickets.
I'll go deeper on counters in the companion piece—but I wouldn't ship an NSM without at least one guardrail.
Once the outcome is clear, I turn it into SMART goals for product managers and test meaningful changes with A/B testing for product managers rather than celebrating a dashboard bump alone.
Using the NSM week to week
- Review it in the same ritual every week (same definition, same window)
- Tie experiments to a hypothesis about the NSM and counters
- Explain movement in plain English to stakeholders
- Revisit annually—or when the product strategy actually changes
My take
A north star metric isn't a slogan. It's a prioritization tool. Pick the outcome that means customers got value, measure it cleanly, and refuse to celebrate lifts that break the product. For outcome context, I pair that choice with product-market fit for product managers and, when loyalty is the goal, jobs to be done for product managers.
If you are still building the measurement habit, start with product analytics for beginners to learn events, funnels, and retention before over-optimizing a single north star.
Want sharper PM analytics judgment? The Product Manager Certification covers the craft—and the newsletter keeps metric field notes coming.
