north-star-metrics

North Star Metrics

Overview

A North Star metric is the single metric that best captures the value your product delivers to users. It aligns the entire organization around a shared goal, prevents sub-optimization, and guides prioritization. This skill helps you define, validate, and operationalize a North Star metric.

When to Use

• Defining the primary success metric for your product
• Aligning the team around a shared goal
• Replacing conflicting or confusing metrics
• You've outgrown vanity metrics and need a real North Star
• Don't use for: short-term campaign measurement, evaluating a single feature, a product with multiple distinct use cases (use a metric tree instead)

Instructions

1. Understand What Makes a Good North Star

A North Star metric should be:

• Customer value-focused: Measures value delivered, not value captured
• Leading indicator: Predicts long-term business success
• Actionable: Teams can directly influence it
• Understandable: Anyone in the company can explain it
• Resistant to gaming: Hard to improve without delivering real value

2. Identify Candidates

Brainstorm 3-5 candidates. Common types:

• Engagement: Daily Active Users, Weekly Active Teams, Messages Sent
• Outcome: Time-to-Value, Tasks Completed, Content Created
• Retention: Day 7 Retention, Active User Rate
• Revenue: Average Revenue Per User (only if tightly linked to value)

For each candidate, ask:

• If this metric goes up, does it mean users are getting more value?
• If this metric goes up, will the business succeed?
• Can every team influence this, even indirectly?

3. Evaluate Against Counter-Metrics

For every North Star candidate, define:

• What could go up while the North Star goes up, but is actually bad?
• Example: Messages Sent goes up, but message quality goes down and spam increases
• These are your counter-metrics — you MUST track them alongside the North Star

4. Test with Real Data

Before committing:

• Pull 6 months of historical data for candidates
• Check correlation with business outcomes (revenue, retention, LTV)
• Test sensitivity: Does the metric move when you ship something good?
• Check volatility: Is it too noisy to be useful for weekly decisions?