name: success-metrics-definer
description: Takes a PRD's success metrics and adds: baseline (where you are now), target (where you want to go), and measurement plan (how you'll track it). Ensures metrics are SMART: specific, measurable, achievable, relevant, time-bound. Returns a metrics validation report.
allowed-tools: Read, Write
effort: medium
Success Metrics Definer
When to activate
Before the PRD is finalized. Use this to turn vague success metrics into measurable, trackable targets.
When NOT to use
Do not use if you don't have access to current baseline data. Baseline is critical; if you can't measure it now, the metric is not valid.
Instructions
- Read the PRD's success metrics section.
- For each metric, define:
- Baseline: Where is this metric right now? (e.g., "Avg onboarding time is currently 45 min")
- Target: Where do you want it after the feature ships? (e.g., "Reduce to 30 min by end of quarter")
- Measurement plan: How will you track it? (e.g., "Track via analytics event 'onboarding_complete' + timestamp")
- Confidence: How certain are you this change is caused by your feature, not external factors? (0–100%)
- Check for SMART criteria:
- Specific: "Increase DAU" is vague; "increase DAU from 50k to 56k" is specific.
- Measurable: defined baseline + target + measurement method.
- Achievable: is a 12% lift realistic given industry benchmarks?
- Relevant: does this metric tie to business outcome (revenue, retention, NPS)?
- Time-bound: by when should the target be achieved?
- Flag any metrics that can't be measured or don't have a baseline.
Output Format
# Success Metrics — [Feature Name]
## Validated Metrics
**Metric 1: [Name]**
- Baseline: [current value + date] | Target: [goal value] | Timeline: [by when]
- Measurement plan: [how we measure; tool/event/report]
- Confidence: [0–100]% that feature causes this change
- SMART check: ✓ (Specific, Measurable, Achievable, Relevant, Time-bound)
---
**Metric 2: [Name]**
- Baseline: [current value] | Target: [goal] | Timeline: [by when]
- Measurement plan: [method]
- Confidence: [0–100]%
- SMART check: ✓
---
## Flagged Metrics (Needs Definition)
**Original:** "Improve user satisfaction"
**Issue:** Vague; no baseline or measurement plan
**Suggested:** "Increase NPS from [current score] to [target score] as measured by [survey frequency]"
---
## Summary
[X] metrics approved | [Y] metrics flagged for revision
Overall readiness: [Ready to launch / Needs refinement]