Leverage Points

---
name: leverage-points
description: Decide WHERE to intervene in a system for maximum effect using Donella Meadows' leverage-points lens — push interventions toward high-leverage places (goals, rules, self-organization, paradigm) and away from low-leverage ones (parameters, buffers, taxes/subsidies). Use when choosing where to act on a complex system, prioritizing competing fixes, designing a policy/intervention/process change, diagnosing why a problem keeps recurring despite effort, or when a past fix didn't stick. The systems-science sharpening of "fix the system, not the symptom."
---

# Leverage Points — where to intervene in a system

Most effort goes to the *lowest*-leverage places (tweak a number, add a buffer) because they're obvious and safe. The highest-leverage places (the system's goal, its rules, its governing paradigm) are less obvious and more resisted — which is exactly why they're high-leverage. This skill is a decision lens for **moving your intervention up the leverage ladder.**

## The core move

When you're about to act on a system, ask: *"What's the highest-leverage place I can realistically intervene here?"* — then push one rung higher than your first instinct.

Meadows' 12 leverage points, **lowest → highest** effect:

| # | Leverage point | Low/High |
|---|---|---|
| 12 | Constants, parameters, numbers (the "tweak a dial" fixes) | lowest |
| 11 | Sizes of buffers and stabilizing stocks | |
| 10 | Structure of material stocks and flows | |
| 9 | Lengths of delays relative to system change | |
| 8 | Strength of negative (balancing) feedback loops | |
| 7 | Gain of positive (reinforcing) feedback loops | |
| 6 | Structure of information flows (who has access to what) | |
| 5 | Rules of the system (incentives, punishments, constraints) | |
| 4 | Power to add, change, or self-organize system structure | |
| 3 | Goals of the system | |
| 2 | Mindset/paradigm the system arises from | |
| 1 | Power to transcend paradigms | highest |

Full descriptions + examples: [`references/meadows-leverage-points.md`](references/meadows-leverage-points.md).

## How to apply (the workflow)

1. **Name the system and the symptom.** What's actually misbehaving, and what's the recurring pain?
2. **Locate your default fix on the ladder.** Most first instincts are #12–#10 (change a number, add a buffer). That's fine to note — it's your baseline.
3. **Climb one or two rungs.** Could you change an *information flow* (#6 — make the invisible visible), a *rule* (#5), or *who's allowed to restructure* (#4) instead of just a parameter? Often a small change at #6/#5 beats a big push at #12.
4. **Check the goal (#3) and paradigm (#2).** When a fix never sticks, the system's *goal* is usually fighting you — you're optimizing a number while the system is steered toward a different end. Surfacing the real goal is frequently the actual intervention.
5. **Pick the highest rung you can realistically move,** and state explicitly which leverage point you're acting on and why.

## Counter-intuitions worth remembering

- **High-leverage points are often pushed in the wrong direction.** People find them by instinct, then turn the dial the wrong way (e.g. adding reinforcing feedback where they needed balancing). Higher leverage = higher blast radius; verify direction before force.
- **Parameters rarely change behavior.** #12 fixes feel productive and almost never alter the pattern. If you've tuned the same number twice, climb the ladder.
- **Information flows (#6) are the cheapest high-leverage move.** Giving an actor feedback they previously lacked changes behavior without changing rules or structure. Look here first when "rules" feel too heavy.

## Worked example

*Symptom:* a recurring carry that never gets done. *Default fix (#12):* re-prioritize it again tomorrow (a parameter tweak — and it's failed twice). *Climb:* #6 — make *why it's carried* visible next to the carry (information flow); #3 — question whether the underlying goal is even still active. The real intervention isn't "try harder tomorrow" (#12) but "surface the reason or kill the goal" (#6/#3).

## Relationship to the rest of the system

This is the systems-science backing for the antifragile principle *"fix the system, not the symptom"* and the *"Systems Over Goals"* operating principle. When `/aios:drift` or `/aios:challenge` surface a stuck pattern, use this lens to choose the intervention altitude.