# Intervention Sizing

> Sizes recommendations for business, data, process, organizational, governance, or technology issues as quick fixes, controlled pilots, structural changes, or transformations. Use when moving from diagnosis to action while avoiding oversized solutions.

- Skill: `bkjohn2018/intervention-sizing` (Agent Skill)
- Install (CLI): `npx skillmds@latest add bkjohn2018/intervention-sizing`
- Raw SKILL.md: https://api.skillmd.com/api/skills/bkjohn2018/intervention-sizing/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: bkjohn2018 (https://skillmd.com/u/bkjohn2018)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/bkjohn2018/intervention-sizing

---


# Intervention Sizing

## When to Use
- A diagnosis has identified likely stressed edges or root-cause hypotheses.
- The user needs recommendations but the right scale of change is uncertain.
- A small test may be better than a large redesign.

## Intervention Sizes
- **Quick fix**: Low-risk clarification, correction, communication, or control adjustment.
- **Controlled pilot**: Small test with measurable feedback and bounded risk.
- **Structural change**: Workflow, ownership, governance, system, data, policy, or control redesign.
- **Transformation**: Broad operating model, platform, culture, funding, or strategy change.

## Workflow
1. Link each proposed action to a symptom and stressed edge.
2. Identify the smallest intervention that can create useful learning.
3. Size the action by blast radius, reversibility, cost, stakeholder impact, and evidence confidence.
4. Define the intended behavior change.
5. Assign owner, first step, success evidence, and feedback mechanism.
6. Escalate only when smaller interventions cannot address the system constraint.

## Output Template
```markdown
## Intervention Sizing
| Recommendation | Size | Symptom addressed | Stressed edge | Intended behavior change | Owner | Feedback mechanism |
|---|---|---|---|---|---|---|
```

## Quality Checklist
- The recommendation is not larger than the evidence supports.
- Quick fixes are low-risk and reversible.
- Pilots have measurable feedback.
- Structural changes explain why edge repair alone is insufficient.
- Transformations are reserved for broad, persistent, high-impact constraints.