# Measure Outcomes

> Defines outcome measures when asked how to measure success, choose a KPI, or distinguish activity from impact; specifies population, units, baseline, and decision use.

- Skill: `jovanipink/measure-outcomes` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx get-skill add jovanipink/measure-outcomes`
- Raw SKILL.md: https://api.skillmd.com/api/skills/jovanipink/measure-outcomes/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: JovaniPink (https://skillmd.com/u/jovanipink)
- Updated: 2026-11-15
- Page: https://skillmd.com/skills/jovanipink/measure-outcomes

---


# KPI and Outcome Measurement

Use generic terminology and preserve the status of every material statement: observed fact, proposal, ratified decision, rejected decision, unresolved question, measured result, estimate, or causal claim.

## Workflow

1. Define the decision the measurement must inform and the outcome it represents.
2. Specify population, unit, numerator, denominator, exclusions, aggregation, time window, and segmentation.
3. Identify source systems, authoritative fields, transformations, latency, missingness, and ownership.
4. Record baseline, target, threshold, confidence, seasonality, and comparison method.
5. Distinguish leading indicators, lagging outcomes, guardrails, quality measures, and operational health.
6. Test incentives, gaming risk, survivorship, selection, attribution, privacy, and subgroup harm.
7. Define review cadence, decision rules, and conditions for retiring or revising the metric.

## Boundaries

- Do not call activity or output an outcome without an explicit causal model.
- Do not invent baselines, targets, or data quality.
- Do not expose personal or sensitive data merely to increase measurement detail.

## Output

Lead with the answer. Use the shortest format that retains the required evidence and limits; expand when requested or needed. The items below describe required content, not mandatory headings.

Return Decision use, Metric dictionary, Data lineage, Baseline and target, Guardrails, Bias and gaming risks, and Review cadence.

## Example

Synthetic example; not an observed client result.

Input: A team counts tasks closed and calls that customer success.

Output: Tasks closed measures output. Define a customer outcome, population, baseline, and observation window before claiming customer improvement.

