Metric Definition Guide
When to Use This Skill
Use this skill when you need to:
- Define how each business metric is calculated and tracked
- Create a metric glossary for team alignment
- Resolve disagreements about what a number means
- Onboard new team members on how to interpret key KPIs
DO NOT use this skill for building dashboards, analyzing data, or setting targets. This is for defining what metrics mean and how they are calculated.
Core Principle
IF TWO PEOPLE IN YOUR COMPANY CALCULATE THE SAME METRIC DIFFERENTLY, YOU DO NOT HAVE A METRIC — YOU HAVE CONFUSION. DEFINE ONCE, USE EVERYWHERE.
Phase 1: Brief
Required Inputs
| Input |
What to Ask |
Default |
| Metrics to define |
"Which KPIs does your team track or should track?" |
Must be provided |
| Business model |
"SaaS, e-commerce, services, marketplace?" |
Must be provided |
| Data sources |
"Where does your data live? (Stripe, GA4, CRM, spreadsheets)" |
Mixed |
| Audience |
"Who will use this guide? (whole team, leadership, new hires)" |
Whole team |
| Known confusion |
"Any metrics your team defines or calculates differently?" |
General misalignment |
GATE: Confirm the metric list before proceeding.
Phase 2: Define
Metric Definition Template
For each metric, document:
- Name — the official metric name (one name, no synonyms in official reporting)
- Definition — one-sentence plain-language explanation
- Formula — exact calculation with numerator and denominator specified
- Data source — where the numbers come from
- Frequency — how often it is calculated (daily, weekly, monthly)
- Owner — who is responsible for tracking and reporting this metric
- Target/benchmark — what "good" looks like
- Interpretation notes — caveats, known quirks, seasonal patterns
GATE: Present the first 3-5 definitions for format approval before completing the full guide.
Phase 3: Build
Deliverables
1. Complete Metric Glossary
- All metrics defined using the standard template
- Organized by category (growth, revenue, engagement, operations)
- Searchable format (table of contents with anchor links)
2. Metric Relationship Map
- How metrics relate to each other (e.g., churn affects LTV which affects CAC:LTV ratio)
- Leading vs. lagging indicator classification
- Which metrics to watch together
3. Calculation Examples
- Worked example for every formula using realistic sample data
- Edge cases: what happens with zero values, negative numbers, or missing data
4. Quick Reference Card
- One-page summary with metric names, formulas, and targets
- Designed for desk reference or Slack pinned message
Phase 4: Polish
Review and Governance
- Review definitions quarterly for accuracy
- When a new metric is proposed, it must go through the definition template before entering any report
- Version control: date-stamp updates and note changes
Onboarding Integration
Include the metric guide in new hire onboarding. Schedule a 30-minute walkthrough of the top 10 metrics during the first week.
Example 1: SaaS Metrics
MRR: Monthly Recurring Revenue. Sum of all active subscription amounts at month end. Source: Stripe. Monthly. Target: 10% MoM growth.
Churn Rate: Customers lost / Customers at start of period. Source: Stripe + CRM. Monthly. Target: Under 3%.
NRR: Net Revenue Retention. (Starting MRR + Expansion - Contraction - Churn) / Starting MRR. Source: Stripe. Monthly. Target: Over 110%.
Example 2: E-commerce Metrics
AOV: Average Order Value. Total Revenue / Number of Orders. Source: Shopify. Weekly. Target: $75+.
Conversion Rate: Orders / Sessions. Source: GA4 + Shopify. Weekly. Target: 2-3%.
ROAS: Return on Ad Spend. Revenue from Ads / Ad Spend. Source: Ad platforms + Shopify. Weekly. Target: 3x+.
Anti-Patterns
- Multiple definitions for the same metric — if marketing calculates "customers" differently than finance, reports will never agree. One definition per metric.
- Undefined denominators — "Conversion rate" means nothing without specifying what divides what over what time period.
- Tribal knowledge — if only one person knows how a metric is calculated, it is not a metric, it is that person's opinion.
- Too many metrics — 50 KPIs means zero focus. Define 8-12 primary metrics and categorize the rest as supporting.
- Never updating definitions — as the business changes, metric definitions must evolve. Review quarterly.
Recovery
- Team cannot agree on a definition: Present 2-3 options with pros and cons. Let the decision-maker pick one and document it as the standard.
- Data sources give different numbers: Identify the discrepancy source. Choose one system of record per metric and document why.
- Too many metrics to define: Start with the 10 metrics that appear in leadership reporting. Add others in batches.
- User unsure which metrics matter: Start with the business model's standard metrics (SaaS: MRR, churn, CAC, LTV; E-commerce: AOV, conversion, ROAS, repeat rate).
1---2name: metric-definition-guide3description: Creates metric glossaries defining how each business KPI is calculated, tracked, and interpreted with formulas, data sources, and ownership. Use for team alignment on numbers.4---56# Metric Definition Guide78## When to Use This Skill910Use this skill when you need to:11- Define how each business metric is calculated and tracked12- Create a metric glossary for team alignment13- Resolve disagreements about what a number means14- Onboard new team members on how to interpret key KPIs1516**DO NOT** use this skill for building dashboards, analyzing data, or setting targets. This is for defining what metrics mean and how they are calculated.1718---1920## Core Principle2122IF TWO PEOPLE IN YOUR COMPANY CALCULATE THE SAME METRIC DIFFERENTLY, YOU DO NOT HAVE A METRIC — YOU HAVE CONFUSION. DEFINE ONCE, USE EVERYWHERE.2324---2526## Phase 1: Brief2728### Required Inputs2930| Input | What to Ask | Default |31|-------|------------|---------|32| **Metrics to define** | "Which KPIs does your team track or should track?" | Must be provided |33| **Business model** | "SaaS, e-commerce, services, marketplace?" | Must be provided |34| **Data sources** | "Where does your data live? (Stripe, GA4, CRM, spreadsheets)" | Mixed |35| **Audience** | "Who will use this guide? (whole team, leadership, new hires)" | Whole team |36| **Known confusion** | "Any metrics your team defines or calculates differently?" | General misalignment |3738**GATE: Confirm the metric list before proceeding.**3940---4142## Phase 2: Define4344### Metric Definition Template4546For each metric, document:47481. **Name** — the official metric name (one name, no synonyms in official reporting)492. **Definition** — one-sentence plain-language explanation503. **Formula** — exact calculation with numerator and denominator specified514. **Data source** — where the numbers come from525. **Frequency** — how often it is calculated (daily, weekly, monthly)536. **Owner** — who is responsible for tracking and reporting this metric547. **Target/benchmark** — what "good" looks like558. **Interpretation notes** — caveats, known quirks, seasonal patterns5657**GATE: Present the first 3-5 definitions for format approval before completing the full guide.**5859---6061## Phase 3: Build6263### Deliverables6465**1. Complete Metric Glossary**66- All metrics defined using the standard template67- Organized by category (growth, revenue, engagement, operations)68- Searchable format (table of contents with anchor links)6970**2. Metric Relationship Map**71- How metrics relate to each other (e.g., churn affects LTV which affects CAC:LTV ratio)72- Leading vs. lagging indicator classification73- Which metrics to watch together7475**3. Calculation Examples**76- Worked example for every formula using realistic sample data77- Edge cases: what happens with zero values, negative numbers, or missing data7879**4. Quick Reference Card**80- One-page summary with metric names, formulas, and targets81- Designed for desk reference or Slack pinned message8283---8485## Phase 4: Polish8687### Review and Governance8889- Review definitions quarterly for accuracy90- When a new metric is proposed, it must go through the definition template before entering any report91- Version control: date-stamp updates and note changes9293### Onboarding Integration9495Include the metric guide in new hire onboarding. Schedule a 30-minute walkthrough of the top 10 metrics during the first week.9697---9899## Example 1: SaaS Metrics100101**MRR:** Monthly Recurring Revenue. Sum of all active subscription amounts at month end. Source: Stripe. Monthly. Target: 10% MoM growth.102**Churn Rate:** Customers lost / Customers at start of period. Source: Stripe + CRM. Monthly. Target: Under 3%.103**NRR:** Net Revenue Retention. (Starting MRR + Expansion - Contraction - Churn) / Starting MRR. Source: Stripe. Monthly. Target: Over 110%.104105## Example 2: E-commerce Metrics106107**AOV:** Average Order Value. Total Revenue / Number of Orders. Source: Shopify. Weekly. Target: $75+.108**Conversion Rate:** Orders / Sessions. Source: GA4 + Shopify. Weekly. Target: 2-3%.109**ROAS:** Return on Ad Spend. Revenue from Ads / Ad Spend. Source: Ad platforms + Shopify. Weekly. Target: 3x+.110111---112113## Anti-Patterns114115- **Multiple definitions for the same metric** — if marketing calculates "customers" differently than finance, reports will never agree. One definition per metric.116- **Undefined denominators** — "Conversion rate" means nothing without specifying what divides what over what time period.117- **Tribal knowledge** — if only one person knows how a metric is calculated, it is not a metric, it is that person's opinion.118- **Too many metrics** — 50 KPIs means zero focus. Define 8-12 primary metrics and categorize the rest as supporting.119- **Never updating definitions** — as the business changes, metric definitions must evolve. Review quarterly.120121---122123## Recovery124125- **Team cannot agree on a definition:** Present 2-3 options with pros and cons. Let the decision-maker pick one and document it as the standard.126- **Data sources give different numbers:** Identify the discrepancy source. Choose one system of record per metric and document why.127- **Too many metrics to define:** Start with the 10 metrics that appear in leadership reporting. Add others in batches.128- **User unsure which metrics matter:** Start with the business model's standard metrics (SaaS: MRR, churn, CAC, LTV; E-commerce: AOV, conversion, ROAS, repeat rate).
reaperinvest