release-metrics

The full skill.

Original instructions from the publisher’s SKILL.md

# Release Metrics

Create a practical measurement plan so product decisions are evidence-based after release.

## When to Use
- Before launch to define success targets.
- After launch to review real performance.
- When deciding whether to iterate, rollback, or expand.

## Defaults
- Minimum metrics: 1 outcome metric and 1 user-experience metric.
- Baseline required before judging success.
- Review cadence: daily for first 3 days, then weekly.

## Procedure
1. Define release objective.
- One sentence objective tied to user value.

2. Choose success metrics.
- Outcome metric examples: task completion rate, activation rate, retention.
- UX metric examples: time-to-complete, error rate, satisfaction score.

3. Define metric formulas.
- Specify numerator, denominator, and data source for each metric.

4. Set baseline and target.
- Baseline from pre-release or pilot data.
- Target with timeframe and confidence assumptions.

5. Add guardrail metrics.
- Track harmful side effects like crashes, auth failures, or support complaints.

6. Define review cadence and owners.
- Who checks each metric and when.
- Trigger thresholds for action.

7. Decide actions from results.
- If target met: scale or expand.
- If near target: iterate focused improvements.
- If below guardrails: hotfix or rollback decision.

## Decision Points
- If no baseline exists: run a short pilot to establish one.
- If data source is unreliable: simplify instrumentation before launch.
- If metrics conflict: prioritize user safety and core outcome metric.

## Completion Criteria
- Objective, metrics, formulas, and data sources are documented.
- Baseline and target are set for each required metric.
- Guardrails and action thresholds are defined.
- Owners and review schedule are assigned.

## Prompt Starters
- /release-metrics Define launch metrics for my first student app release.
- /release-metrics Build a weekly review plan and go/no-go thresholds.