value-metrics

Measure what matters, ignore what doesn't.

How to use

• /value-metrics Apply metrics constraints to this conversation.
• /value-metrics <product-type> Recommend metrics for the described product.

Constraints

North Star Metric

• MUST identify ONE metric that best represents the value users get
• This metric should go up when users are getting value and down when they're not
• Examples: messages sent (Slack), nights booked (Airbnb), deployments completed (Vercel)
• NEVER use revenue as the north star. Revenue is an output, not a driver.
• NEVER use DAU/MAU without defining what counts as a meaningful visit

Metric Hierarchy

1. North Star: the one number everyone aligns on
2. Input metrics: 3-5 metrics that drive the north star
3. Health metrics: things that shouldn't drop (performance, error rate, support volume)
4. Guardrail metrics: things you watch to make sure growth isn't harming quality

Analytics Events

• MUST track the full activation funnel (signup > onboarding step > core action > retained)
• MUST track where users drop off, not just where they convert
• SHOULD track time-to-value (how long from signup to first meaningful outcome)
• NEVER track more than you'll actually look at. Unused data is not an asset.

Dashboard Rules

• MUST be scannable in under 10 seconds
• SHOULD show trends over time, not just current numbers
• MUST include comparison period (vs. last week, last month)
• NEVER show metrics without context (is 1,000 signups good or bad?)
• SHOULD alert on anomalies, not just thresholds

Anti-Patterns

• Measuring what's easy to measure instead of what matters
• Vanity metrics (total signups, page views, social followers)
• Building dashboards nobody checks
• Changing metrics to make bad performance look good
• Measuring activity instead of value (logins are not engagement)