| name | north-star-metric |
| description | Define or critique a product North Star Metric and its input metrics. Use when a team needs one shared leading indicator of customer value and growth. |
| argument-hint | ["product-or-problem-space"] |
North Star Metric
Use this skill when the user wants to define, refine, critique, or operationalize a North Star Metric (NSM) for a product, feature set, or business line.
What this skill must produce
Always produce these sections:
- Context summary
- Candidate engagement game / product motion
- North Star options (2-4 options)
- Recommended North Star
- Input metrics (3-6)
- Metric definition spec
- Why this is a good North Star
- Risks / failure modes
- Instrumentation requirements
- 30/60/90 day rollout plan
If the user asks only for critique, still use the same structure, but focus on diagnosing the existing North Star.
Inputs to gather
If the user or repo already provides them, extract:
- Product name
- Primary user
- Primary customer / buyer
- Core job-to-be-done
- Core product motion: attention, transaction, productivity, or hybrid
- Business model: freemium, self-serve, sales-assisted, enterprise, usage-based, seat-based, subscription
- Core user actions
- Desired business outcomes
- Existing KPIs / dashboards / event schema
- Whether this is company-level, product-line-level, or team-level
If something is missing, make a best-effort assumption and label it clearly.
Working rules
- Prefer a leading indicator over lagging outcomes like revenue, MRR, or ARPU.
- Prefer a metric that captures customer value realized, not vanity activity.
- Prefer a metric that product and marketing can influence through day-to-day work.
- Prefer one metric that is hard to game.
- Avoid metrics that can rise while product value falls.
- Avoid pure volume metrics when a quality threshold is required.
- Avoid making the NSM directly editable by one team without the supporting system beneath it.
Step-by-step method
Step 1: frame the product and value exchange
State in one sentence:
- who the user is
- what they are trying to get done
- what value the product creates
- what value the business gets back
Use this format:
Value exchange: When [user] does [meaningful action] to achieve [outcome], the business gains [business value].
Step 2: identify the engagement game
Classify the product primarily as:
- Attention: value comes from time, content consumption, expression, or repeated engagement
- Transaction: value comes from completing purchases / bookings / exchanges with confidence
- Productivity: value comes from completing a workflow more effectively or reliably
If hybrid, pick the dominant one and explain why.
Step 3: generate candidate North Stars
Create 2-4 candidate NSMs.
For each candidate provide:
- metric name
- formula
- why it may work
- why it may fail
- what behavior it encourages
- what anti-pattern it risks
Step 4: select the recommended NSM
Choose the best candidate and explain the tradeoff.
A recommended NSM should pass this checklist:
- expresses customer value
- represents product strategy
- is a leading indicator
- is actionable
- is understandable
- is measurable
- is not a vanity metric
Step 5: define input metrics
Create 3-6 input metrics that plausibly drive the NSM.
For each input metric specify:
- name
- formula
- owner
- weekly directionality
- why it matters
- what team can influence it
Inputs should be operational levers, not just more outcomes.
Step 6: create the metric definition spec
Write an implementation-ready spec with:
- canonical metric name
- plain-English definition
- numerator
- denominator (if any)
- quality thresholds
- time window
- aggregation level
- exclusions
- segment cuts
- lagging outcomes expected to move if this improves
Step 7: stress-test the metric
Explicitly answer:
- How could this metric be gamed?
- Could it increase while user value falls?
- What seasonal, cohort, or market distortions affect it?
- What would make us revisit it?
Step 8: connect to instrumentation
List the minimum event and property requirements required to compute the NSM and each input metric.
Output format
Use exactly this structure:
Context summary
- Product:
- User:
- Buyer:
- JTBD:
- Motion:
- Business model:
Candidate engagement game / product motion
...
North Star options
| Option | Formula | Why it could work | Main risk |
|---|
Recommended North Star
Name:
Formula:
Interpretation:
Input metrics
| Input metric | Formula | Owner | Why it matters |
|---|
Metric definition spec
- Canonical name:
- Plain-English definition:
- Grain:
- Time window:
- Quality threshold:
- Included events:
- Excluded events:
- Segments to monitor:
Why this is a good North Star
- Expresses customer value:
- Represents strategy:
- Leading indicator:
- Actionable:
- Understandable:
- Measurable:
- Non-vanity:
Risks / failure modes
Instrumentation requirements
- Events:
- Event properties:
- User properties:
- Account / org properties:
- Data quality checks:
30/60/90 day rollout plan
- 30 days:
- 60 days:
- 90 days:
Failure mode handling
If you cannot recommend a single NSM with confidence:
- say that directly
- provide the top 2 candidates
- explain the missing evidence needed to choose between them
- recommend a temporary proxy metric and a validation plan