# Portfolio And Measurement

> Improve existing content and close the learning loop without cannibalizing new-content work.

- Skill: `lionelndong/portfolio-and-measurement` (Agent Skill)
- Install (CLI): `npx skillmds add lionelndong/portfolio-and-measurement`
- Raw SKILL.md: https://api.skillmd.com/api/skills/lionelndong/portfolio-and-measurement/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: lionelndong (https://skillmd.com/u/lionelndong)
- Updated: 2026-08-19
- Page: https://skillmd.com/skills/lionelndong/portfolio-and-measurement

---


# Portfolio and Measurement

## Portfolio review

Run separately from new content. Classify candidates before scoring: live integrity, customer/revenue loss, rank/conversion, link equity, maintenance, or no-op. Within a class, prioritize with `Business Value × Impact × Confidence × Urgency ÷ Effort` and evidence citations.

Select at most one focused action: update, relaunch, merge recommendation, delete recommendation, leave, technical repair, or conversion repair. Never execute merge, delete, redirect, deindex, outreach, or public promotion without explicit approval.

## Measurement

For every shipped change, record baseline, success metric, guardrail metric, and a 14–21 day measurement date. Measure discovery, visibility, qualified visits, activation, conversion, paid outcomes, links, and compounding versus flatlining. If the window is incomplete, record `insufficient_data` and schedule the next measurement; do not leave a run open.

## Output

Candidate register, selection rationale, focused action packet, baseline, measurement record, and keep/iterate/revert recommendation.