07 · Metrics & KPIs Basics¶
A PRD's "success metrics" section (Module 5) only works if you know how to choose good metrics in the first place. This module covers the North Star metric concept and the AARRR ("pirate metrics") framework — two complementary tools for deciding what to measure and why.
North Star metric¶
A North Star metric is the single measure that best captures the core value a product delivers to its users — chosen so that, if it goes up, the business is healthier, and every team can trace their work back to influencing it.
A good North Star metric is a leading indicator of value delivered, not just a lagging business outcome like revenue (which is affected by too many outside factors to guide day-to-day product decisions).
| Product | Weak North Star (lagging/vague) | Better North Star (leading, specific) |
|---|---|---|
| Music streaming app | Total revenue | Weekly hours of music listened per active user |
| Project-management SaaS | Total signups | Number of teams with 3+ active weekly users |
| Ride-sharing app | Total rides | Weekly active riders who complete a ride |
| Photo-editing app | Downloads | Monthly active users who export/share an edited photo |
Notice each "better" version ties directly to the moment the user actually gets value — not just adjacent activity (a signup isn't value delivered; a completed ride is).
AARRR — the "pirate metrics" funnel¶
AARRR breaks the user lifecycle into five stages, giving you a metric to track — and a bottleneck to look for — at each one.
| Stage | Question it answers | Example metric |
|---|---|---|
| Acquisition | How do users find us? | Signups by channel per week |
| Activation | Do new users have a good first experience? | % of signups who complete onboarding / reach "first value" moment |
| Retention | Do users come back? | % of users active in week 2 who were also active in week 1 |
| Referral | Do users bring others? | % of active users who invite at least one teammate |
| Revenue | Do users pay (or convert to paid)? | Free-to-paid conversion rate |
The value of AARRR is diagnostic: if overall growth stalls, this funnel lets you localize where — a product with strong acquisition but weak activation has a fundamentally different problem (and fix) than one with strong activation but weak retention.
Leading vs. lagging metrics¶
| Type | Definition | Example | Use for |
|---|---|---|---|
| Leading | Predicts future outcomes; you can act on it now | Weekly active users, onboarding completion rate | Day-to-day product decisions |
| Lagging | Confirms past outcomes; useful for reporting, less for steering | Quarterly revenue, churn rate (measured after the fact) | Business reporting, board updates |
A common mistake is a team obsessing over a lagging metric (like quarterly churn) week to week — by the time it moves, the decisions that caused the move are months old. Track leading metrics for steering; report lagging metrics for context.
Choosing metrics for a specific initiative¶
Every PRD needs its own success metrics, distinct from (but ideally contributing to) the company's North Star. A simple checklist:
| Question | Why it matters |
|---|---|
| Can we actually measure this with existing instrumentation? | An unmeasurable metric can't confirm success |
| Is there a clear baseline before launch? | Without a baseline, "improvement" is unfalsifiable |
| Does moving this metric plausibly move the North Star? | Prevents optimizing a vanity metric disconnected from real value |
| What's a specific, numeric target (not just "improve")? | "Improve engagement" isn't checkable; "+10% week-2 retention" is |
Worked example¶
A project-management SaaS's North Star is "teams with 3+ weekly active users." The team notices Retention (AARRR) is the weak link: strong signups and activation, but only 35% of teams stay active past week 3. Digging in with interviews (Module 2), they learn teams stop returning once the initial project setup is "done" — there's no ongoing reason to open the tool mid-week. They prioritize (Module 4) a "weekly digest email" initiative whose PRD (Module 5) sets a specific success metric: "+8 percentage points in week-3 team retention within 6 weeks of launch, measured against the current 35% baseline" — a leading, specific, measurable target that plainly ties back to the North Star.
How It Actually Works¶
The reason "vanity metrics" mislead is a specific statistical failure: they're metrics that trend upward under almost any condition (e.g., total signups, page views) because they accumulate monotonically and correlate more with the passage of time and marketing spend than with product quality — they lack a denominator that would reveal rate or ratio, which is where the real signal lives. A north star metric works as an organizational tool because of Goodhart's Law (when a measure becomes a target, it ceases to be a good measure): teams will optimize whatever they're measured on, so the choice of metric silently redirects behavior, meaning the metric selection process is really an incentive-design exercise disguised as an analytics exercise. Leading vs. lagging indicators matter mechanically because of measurement latency — revenue and retention (lagging) take weeks or months to move and reflect decisions made long ago, while activation rate or feature adoption (leading) move within days and are causally closer to the decisions you're making this week, so a team that only watches lagging metrics is flying with a multi-week-delayed instrument panel, reacting to problems that are already too old to cheaply fix.
Exercise¶
For the initiative you've been building a PRD around (Modules 3-6), write:
- A proposed North Star metric for the overall product it belongs to (even if you're inventing the product context) — and one sentence on why it's leading rather than lagging.
- Which AARRR stage your initiative most directly targets, and why.
- A specific, numeric success metric for the initiative itself, including a stated baseline and target (estimate plausibly if you don't have real data, but state your baseline assumption explicitly).