Skip to main content
AI/MLcoalesce-labs

define-north-star

Identify and validate your North Star Metric. Aligns product strategy with key business metric.

Stars
12
Source
coalesce-labs/catalyst
Updated
2026-05-31
Slug
coalesce-labs--catalyst--define-north-star
View on GitHubRaw SKILL.md

// install — copy + paste into any project

mkdir -p .claude/skills && curl -fsSL https://raw.githubusercontent.com/coalesce-labs/catalyst/HEAD/plugins/pm/skills/define-north-star/SKILL.md -o .claude/skills/define-north-star.md

Drops the SKILL.md into .claude/skills/define-north-star.md. Works with Claude Code, Cursor, and any agent that loads SKILL.md files from .claude/skills/.

Define North Star Metric

When to use: During strategy planning, when metrics feel scattered, or when teams are optimizing different things

Framework source: Aakash Gupta's "Do you really need a North Star Metric?"

Quick Start

  1. Tell me: "Help me define our North Star metric" (or "Validate our current North Star")
  2. I will check [[business-info-template]] and thoughts/shared/pm/frameworks/ for your business model, growth stage, and existing metrics
  3. I will ask about your core value, retention drivers, and business model to narrow candidates
  4. We work through: Core Value identification, Metric formula (Frequency x Core Action x Breadth), Validation tests, Input metrics, and Guardrails
  5. Output goes to thoughts/shared/pm/metrics/north-star-[quarter].md

Key decision: Not every product needs a single North Star. Marketplaces, multi-product companies, and complex B2B may need a constellation of 2-4 metrics instead. I will help you decide which approach fits.

Context Routing Logic (Internal - for Claude)

Automatic Context Checks: When this skill is invoked, immediately check:

Source Files/Folders Search Terms What to Extract
Strategy Docs thoughts/shared/pm/frameworks/*.md objective, business goal, success metric Current metric direction, if any
Business Model thoughts/shared/pm/context/business-info-template.md revenue model, growth focus, metrics What drives the business
Metrics History thoughts/shared/pm/metrics/*.md baseline, trends, retention data Current metric baselines and movement
Meetings thoughts/shared/product/meeting-notes/*.md "North Star", "KPI", "success metric" Stakeholder expectations
PRDs thoughts/shared/pm/prds/*.md success metric, target Feature-level success indicators

Context Priority:

  1. Business model and revenue drivers FIRST
  2. Current product stage and growth focus SECOND
  3. Historical metrics data THIRD
  4. Stakeholder expectations FOURTH

Cross-Skill Links:

  • If building strategy → Link to /write-prod-strategy which uses North Star
  • If defining feature metrics → Link to /feature-metrics which should ladder to North Star
  • If analyzing retention → Link to /retention-analysis to identify leading indicators
  • If setting up metrics framework → Link to /metrics-framework

Step 0: Understanding Your Current Metric State

Before defining your North Star, let me understand where you are...

Checking:

  • thoughts/shared/pm/context/business-info-template.md for your business model
  • thoughts/shared/pm/frameworks/ for strategic direction
  • thoughts/shared/pm/metrics/ for baseline metrics and trends
  • thoughts/shared/product/meeting-notes/ for stakeholder priorities

Based on what I find, I'll show you:

Current Metric State

Business Model:

  • [How do you make money? Subscription / marketplace / ads / transactions?]
  • [Growth stage: acquisition focus / retention focus / monetization focus?]
  • [TAM: addressable market size]

Existing Metrics:

  • [Current metrics dashboard: what are you tracking?]
  • [Metric trends: what's moving, what's flat?]
  • [Stakeholder focus: what does leadership care about?]

Metric Gaps:

  • [Metrics that aren't connecting to decisions: vanity metrics?]
  • [Decisions being made without data: where are you flying blind?]

PM-Specific Diagnosis Questions

  1. Business Model Clarity: Are you clear on what drives revenue/value in your business?
  2. Growth Focus: Are you optimizing for growth, engagement, retention, or monetization?
  3. Team Alignment: Do all teams agree on what success looks like?
  4. Data Quality: Do you trust your existing metrics data?
  5. Decision Speed: How quickly do you need to make decisions? (Hourly, daily, weekly, monthly?)

What is a North Star Metric?

Definition: The single metric that best captures the core value your product delivers to customers.

Do You Actually Need One?

When a North Star Metric Works Well:

✅ Product-led growth (PLG) companies

  • Self-serve product
  • Clear activation moment
  • Example: Slack's "Teams sending 2,000+ messages"

✅ Consumer products with network effects

  • Value increases with usage
  • Example: Facebook's "Daily Active Users"

✅ Early-stage companies (Seed to Series B)

  • Need focus and alignment
  • Limited resources require prioritization

When a North Star Can Be Misleading:

❌ Marketplace/multi-sided platforms

  • Optimizing for buyers might hurt sellers
  • Need balanced metrics for both sides
  • Example: Airbnb tracks host AND guest metrics

❌ Complex B2B enterprise products

  • Multiple stakeholders with different needs
  • Example: Salesforce has different success metrics by user role

❌ Companies with multiple business models

  • Each line of business may need different metrics
  • Example: Amazon (retail vs AWS vs ads)

The North Star Framework

A good North Star Metric has three components:

1. Core Action

What is the one thing users do that indicates they're getting value?

Examples:

  • Spotify: Time spent listening
  • Airbnb: Nights booked
  • Slack: Messages sent
  • LinkedIn: Weekly active users
  • Notion: Documents created

2. Frequency

How often should this action happen to indicate real value?

Examples:

  • Daily: Social apps (Instagram, TikTok)
  • Weekly: Productivity tools (Notion, design tool)
  • Monthly: Marketplace (Airbnb, Uber)
  • Per-session: E-commerce (Amazon)

3. Breadth

Who needs to do this action for it to count?

Examples:

  • Individual users: Netflix (hours watched per user)
  • Teams: Slack (active teams sending messages)
  • Multi-sided: Uber (completed trips)

Formula: [Frequency] × [Core Action] × [Breadth]

Examples:

Slack:

  • Frequency: Daily
  • Core Action: Messages sent
  • Breadth: Active teams
  • North Star: Daily Active Teams sending 2,000+ messages

Airbnb:

  • Frequency: Per transaction
  • Core Action: Nights booked
  • Breadth: Guests
  • North Star: Nights booked

Spotify:

  • Frequency: Weekly
  • Core Action: Time listening
  • Breadth: Active users
  • North Star: Weekly Active Users × Hours listened

How to Find Your North Star

Step 1: Identify Your Core Value

Answer these questions:

  1. What problem does your product solve?

    • Not features, but the actual problem

  2. When do users say "aha, this is valuable"?

    • Look at retention data: what do retained users do differently?

  3. What action most correlates with retention?

    • Run cohort analysis
    • Compare churned vs retained users

Use this prompt:

Use /define-north-star and reference [[business-info-template]]

Help me identify our core value:
- Product: [describe your product]
- Customer problem: [what problem you solve]
- Retention data: [describe retained vs churned user behavior]

What action best captures when users get value?

Step 2: Validate Against Criteria

Your North Star should be:

✅ Measurable

  • Can you track it accurately?
  • Do you have the data infrastructure?

✅ Actionable

  • Can teams influence it with their work?
  • Are there levers to pull?

✅ Leading indicator

  • Does it predict revenue/retention?
  • Or is it just a vanity metric?

✅ Understandable

  • Can everyone in the company explain it?
  • Does it pass the "grandma test"?

✅ Captures sustainable value

  • Not gameable
  • Reflects real customer value

Step 3: Test for Tradeoffs

The real test: Does it help you make hard decisions?

Example decision scenarios:

Scenario 1: You can either:

  • A) Improve conversion rate +10% (more signups)
  • B) Improve Day 7 retention +5% (better activation)

Which moves your North Star more?

Scenario 2: Marketing wants to:

  • Run a campaign that brings 10K new users (low quality)
  • Or 1K highly-targeted users (high quality)

Which aligns with your North Star?

If your North Star doesn't help you choose, it's not working.

Common North Star Metrics by Category

Social / Community Products

  • Facebook: Daily Active Users (DAU)
  • Instagram: Daily Active Users sharing stories
  • TikTok: Time spent watching videos
  • Discord: Weekly Active Servers

Productivity / Collaboration

  • Slack: Daily Active Teams (2,000+ messages)
  • Notion: Weekly Active Users creating docs
  • design tool: Weekly Active Designers
  • Asana: Tasks completed per week

Marketplace / Transaction

  • Airbnb: Nights booked
  • Uber: Completed trips
  • Etsy: Gross Merchandise Value (GMV)
  • DoorDash: Orders completed

Media / Content

  • Netflix: Hours watched per member
  • Spotify: Hours listened per subscriber
  • Medium: Time reading articles
  • YouTube: Watch time

SaaS / B2B

  • Salesforce: Active users per account
  • HubSpot: Weekly active contacts managed
  • Stripe: Payment volume processed
  • Dropbox: Active files shared

Red Flags: Bad North Star Metrics

❌ Revenue as North Star

Problem: Revenue is an outcome, not a value metric

  • Doesn't tell you WHY customers pay
  • Can go up even if product value decreases (price increases)
  • Lagging indicator (slow to respond)

Exception: Marketplaces where GMV = value delivered

❌ Number of Features Used

Problem: More features ≠ more value

  • Can be gamed by adding trivial features
  • Doesn't capture depth of value

Better: Usage of core feature

❌ Vanity Metrics (Signups, Downloads, Page Views)

Problem: No correlation to actual value or retention

  • Can grow while business dies
  • Doesn't reflect customer success

Better: Activation rate, active users

From North Star to Execution

Once you have your North Star, break it down into input metrics:

Example: Airbnb's "Nights Booked"

Input Metrics (things that drive Nights Booked):

  1. Supply side:

    • Active listings
    • Listing quality score
    • Host responsiveness

  2. Demand side:

    • Search-to-booking conversion
    • Repeat booking rate
    • Average trip length

  3. Trust & safety:

    • Review ratings
    • Dispute rate
    • Insurance claims

Each team owns input metrics that ladder up to North Star.

How Teams Use the North Star

Product Team:

"Will this feature increase Nights Booked?"

  • If yes → prioritize
  • If no → deprioritize

Marketing Team:

"Will this campaign bring users who book nights?"

  • Focus on qualified traffic, not just signups

Engineering Team:

"Will improving performance increase Nights Booked?"

  • Faster search → more bookings → prioritize

Customer Success:

"Are our customers booking nights regularly?"

  • If not, something's wrong with value delivery

Evolving Your North Star

Your North Star should evolve as your company matures:

Early stage (0-100K users):

  • Focus: Activation and retention
  • North Star: Weekly Active Users

Growth stage (100K-1M users):

  • Focus: Sustainable growth loops
  • North Star: Weekly Active Users engaging with core feature

Scale stage (1M+ users):

  • Focus: Monetization and expansion
  • North Star: Revenue per active user OR engagement + monetization hybrid

Enterprise stage:

  • Multiple North Stars for different business units

Real-World Example: Slack

Slack's North Star Evolution:

2014-2015 (Early):

  • North Star: Teams sending 2,000+ messages
  • Why: Indicated real team adoption

2016-2018 (Growth):

  • North Star: Daily Active Teams (2,000+ messages)
  • Why: Added frequency dimension

2019-2021 (Scale):

  • North Star: Weekly Active Users + Messages sent
  • Why: Balance growth with engagement

2022+ (Enterprise):

  • North Star: Active paid seats
  • Why: Focus on monetization at scale

Key insight: North Star evolved with company stage

Worksheet: Define Your North Star

1. Core Value

What problem do you solve? **___**

2. Core Action

What do users do when they get value? **___**

3. Retention Correlation

What action correlates most with retention? **___**

4. Frequency

How often should users do this action? **___**

5. Breadth

Who needs to do this? (Individual, team, transaction) **___**

6. North Star Candidate

Formula: [Frequency] × [Core Action] × [Breadth]

7. Validation Checks

  • Measurable? (Can we track it?)
  • Actionable? (Can teams influence it?)
  • Leading indicator? (Predicts retention/revenue?)
  • Understandable? (Everyone gets it?)
  • Sustainable? (Reflects real value?)

8. Tradeoff Test

Describe 2-3 hard decisions your North Star should help you make:

Alternative: Constellation of Metrics

If a single North Star doesn't work for you, use a constellation:

Example: Amazon

  • Customer satisfaction (NPS)
  • Selection (# of products)
  • Price (value for money)
  • Convenience (delivery speed)

Rule: No more than 4 metrics in your constellation

  • More than 4 = unfocused strategy

Guardrail Metric Guidance

Every North Star needs 2-3 guardrail metrics to prevent destructive optimization.

Without guardrails, teams will find ways to game the North Star that hurt the business. The question to ask: "If we aggressively optimized [North Star], what could go wrong? That is your guardrail."

Common guardrail patterns:

  1. Quality metric (e.g., NPS, CSAT, support ticket volume) -- Prevents growth at the expense of user satisfaction. Example: If North Star is "Weekly Active Users," a guardrail of "NPS stays above 40" prevents acquiring low-quality users who churn immediately.

  2. Revenue/monetization metric (e.g., ARPU, conversion rate, LTV) -- Prevents engagement optimization that does not generate revenue. Example: If North Star is "time spent in app," a guardrail of "ARPU stays above $12/mo" prevents building addictive features that do not convert to paid.

  3. Efficiency metric (e.g., CAC, cost per transaction, infrastructure cost per user) -- Prevents growth that is not sustainable. Example: If North Star is "completed transactions," a guardrail of "CAC stays below $50" prevents buying growth that does not pay back.

How to define guardrails:

  • Set a minimum threshold (floor), not a target. Guardrails are "do not go below X," not "try to reach Y."
  • Define what happens if breached: who gets alerted, what investigation to run, and what corrective action to take.
  • Review guardrails quarterly. As the product matures, guardrail thresholds may need adjustment.

Output Template

North Star Metric Analysis: [Company Name]

Recommended North Star: [Metric Name] Formula: [How it is calculated] Current Value: [Baseline] Target: [Goal + timeline]

Validation

  • Core Value Test: [Pass/Fail + reasoning]
  • Leading Indicator Test: [Pass/Fail + reasoning]
  • Tradeoff Test: [Pass/Fail + reasoning]

Input Metrics (Driver Tree)

Input Metric Current Target Owner Leading Indicator For
[Metric 1] [Value] [Goal] [Team/Person] [What it predicts]
[Metric 2] [Value] [Goal] [Team/Person] [What it predicts]
[Metric 3] [Value] [Goal] [Team/Person] [What it predicts]

Guardrail Metrics

Guardrail Why Threshold Action if Breached
[Quality metric] [Prevents what failure mode] [Do not drop below X] [Who investigates + what they do]
[Revenue metric] [Prevents what failure mode] [Do not drop below X] [Who investigates + what they do]
[Efficiency metric] [Prevents what failure mode] [Do not drop below X] [Who investigates + what they do]

Recommendation

[2-3 sentences on whether to adopt/keep/change the North Star, with timeline for review. Example: "Adopt Weekly Active Teams as your North Star for the next 6 months. Review at the end of Q3 to determine if the metric still reflects core value delivery as you move into enterprise. Schedule a North Star review meeting for September."]

Common Mistakes

❌ Changing your North Star every quarter

  • Causes whiplash across teams
  • Choose one and stick with it for 6-12 months

❌ Having different North Stars by team

  • Defeats the purpose of alignment
  • Everyone should rally around one metric

❌ Making it too complex

  • If it requires a 10-minute explanation, it's too complex
  • Should fit in one sentence

❌ Ignoring guardrails

  • Optimize for North Star, but don't break the product
  • Always have 3-5 guardrail metrics

Output Integration

Where Files Go

North Star definition:

  • Save to: thoughts/shared/pm/metrics/north-star-[quarter].md (your working definition)
  • Share widely: This becomes the basis for all team metrics

Link to Other Work

After defining North Star:

  • Reference in strategy - "Our North Star metric is [X]" in /write-prod-strategy
  • Guide PRDs - All feature PRDs should explain how they support North Star
  • Inform roadmap - Roadmap priorities are evaluated by North Star impact
  • Dashboard anchor - North Star sits at top of your metrics dashboard

Cross-Skill Integration

Feeds into:

  • /write-prod-strategy - North Star becomes the Objective component
  • /metrics-framework - North Star is your lagging metric anchor
  • /feature-metrics - Feature success metrics should ladder to North Star
  • /catalyst-pm-ops:status-update - Progress toward North Star is tracked in updates

Pulls from:

  • [[business-info-template]] - Business model informs what matters
  • thoughts/shared/pm/metrics/ - Historical data validates metric choices
  • /retention-analysis - Understand what drives long-term success
  • /activation-analysis - Early indicators of North Star movement

Related Skills

  • /activation-analysis - Find your activation metric
  • /feature-metrics - Choose experiment metrics that ladder to North Star
  • /metrics-framework - Understand leading vs lagging metrics
  • /retention-analysis - Measure what drives retention and North Star
  • /write-prod-strategy - Connect North Star to strategy

Output Quality Self-Check

Before delivering the North Star analysis, verify:

  • Metric is one sentence: The North Star can be stated in a single clear sentence (e.g., "Weekly Active Teams sending 2,000+ messages"). If it needs a paragraph to explain, it is too complex.
  • Formula is explicit: The calculation is defined precisely enough that an analyst could implement it in SQL. No ambiguity about what counts and what does not.
  • Baseline and target are included: Current value and target value with timeline are both specified. "Improve engagement" is not a North Star.
  • Three validation tests passed: Core Value Test, Leading Indicator Test, and Tradeoff Test are each explicitly addressed with pass/fail and reasoning.
  • Input metrics defined: At least 3 input metrics are identified with owners, showing how teams can influence the North Star.
  • Guardrails defined: 2-3 guardrail metrics are specified with thresholds and breach actions. A North Star without guardrails is dangerous.
  • Tradeoff scenario tested: At least one real decision scenario is described where the North Star helps choose between two options. If it does not help decide, it is not working.
  • Review timeline set: A specific date to revisit the North Star is recommended (typically 6-12 months).

Framework credit: Adapted from Aakash Gupta's North Star Metric framework. Read the full article: https://www.news.aakashg.com/p/do-you-really-need-a-north-star-metric