Find Leverage Points Command

You are analyzing intervention options using Donella Meadows' 12 places to intervene in a system. Your goal is to find the highest-leverage point that's feasible given constraints.

Core Principle

Small shifts at high leverage points beat massive efforts at low leverage points.

Most people intervene at Level 12 (parameters) because it's obvious and easy. Systems thinkers intervene at Levels 6-3 (information, rules, goals) because that's where change actually happens.

The 12 Places to Intervene (Weakest to Strongest)

Low Leverage (12-10): Low Risk, Low Reward

12. Constants, Parameters, Numbers

• What: Changing quantities (budgets, quotas, targets)
• Example: "Add more servers", "Hire more engineers"
• Why weak: Structure stays intact, system adapts

11. Buffers (Stabilizing Stocks)

• What: Reserve capacity that absorbs fluctuation
• Example: Connection pool size, retry queues
• Why stronger: Prevents cascade failures

10. Stock-and-Flow Structures

• What: Physical plumbing, topology
• Example: Microservices vs monolith
• Why stronger: Changes what's possible

Medium Leverage (9-7): Moderate Risk and Reward

9. Delays

• What: Time between action and consequence
• Example: CI/CD speed, monitoring latency
• Warning: Shortening delays in reinforcing loops accelerates problems

8. Balancing Feedback Loops

• What: Error-correction mechanisms
• Example: Auto-rollback, rate limiters
• Use when: Want stability and self-correction

7. Reinforcing Feedback Loops

• What: Amplification mechanisms
• Example: Network effects, technical debt spiral
• Use when: Want to amplify virtuous cycles, dampen vicious ones

High Leverage (6-5): High Reward, Moderate-High Risk

6. Information Flows

• What: Who sees what information when
• Example: Real-time dashboards, transparent incident reports
• Why counterintuitive: Seems passive but often more powerful than mandates

5. Rules

• What: Incentives, constraints, permissions
• Example: "You build it, you run it", deployment windows
• Warning: Rules get gamed if goals misaligned

Highest Leverage (4-1): Highest Reward, Highest Risk

4. Self-Organization

• What: System's ability to evolve its own structure
• Example: Engineer-driven RFC process, autonomous teams
• Risk: May optimize locally at expense of global

3. Goals

• What: System purpose
• Example: "Prevent incidents" → "Learn from incidents"
• Why powerful: Everything else serves the goal

2. Paradigms

• What: Mental models, worldview
• Example: "Engineers as resources" → "Engineers as investors"
• Why powerful: Changes how we see everything

1. Transcending Paradigms

• What: Ability to step outside any paradigm
• Example: "Strong opinions, weakly held"
• Why powerful: Not attached to any one way of seeing

Analysis Process

Step 1: Identify Current Level

What level is your proposed solution?

If your solution...	Level
Adjusts a number, budget, quantity	12
Adds capacity, reserves, slack	11
Redesigns architecture, topology	10
Speeds up or slows down a process	9
Adds monitoring, alerts, auto-scaling	8
Amplifies growth or dampens decline	7
Makes something visible, adds transparency	6
Changes policies, mandates, incentives	5
Enables teams to self-organize	4
Redefines what success means	3
Changes fundamental assumptions	2
Questions whether problem is real	1

Step 2: Generate Higher-Level Alternatives

Ask "Why?" three times, then intervene there:

Example: "We need more servers"

• Why? Response time is slow
• Why? 20 serial service calls
• Why? Designed for sync everywhere
• Intervention: Question "sync by default" paradigm (Level 2)

Move up systematically:

For solution at Level N, ask:

• Level N+1: What rule would make this parameter self-adjust?
• Level N+2: What information would make people want this?
• Level N+3: What goal would make this rule unnecessary?
• Level N+4: What paradigm shift would make this goal obvious?

Step 3: Assess Prerequisites

Higher leverage requires more prerequisites:

Level	Prerequisites Required
12-10	None, safe to experiment
9-7	Map system structure first
6-5	Leadership buy-in, understand power structures
4-1	Psychological safety, organizational readiness, patience

Step 4: Choose Appropriate Level

Consider:

• Urgency (lower levels are faster)
• Sustainability (higher levels last longer)
• Prerequisites available
• Resistance expected

Often best: Multi-level approach

• Level 12 tactically (buy time)
• Higher level strategically (sustainable change)

Output Format

# Leverage Analysis: [Problem/Solution]

## Current Proposal
**Solution:** [Description]
**Level:** [#] - [Level name]

## Higher-Level Alternatives

### Level [N+1]: [Level name]
**Alternative:** [Description]
**Prerequisite:** [What's needed]
**Expected resistance:** [From whom]

### Level [N+2]: [Level name]
[...]

## Recommendation

**Tactical (immediate):** Level [#] - [Description]
**Strategic (sustainable):** Level [#] - [Description]

**Rationale:** [Why this combination]

## Prerequisites to Address
- [Prerequisite 1]
- [Prerequisite 2]

Red Flags - Rationalizations for Low Leverage

Thought	Response
"Too urgent for high-leverage thinking"	Urgency is when leverage matters most
"High-leverage is too slow"	Low-leverage that fails is slower
"High-leverage is too risky"	Repeating failed low-leverage is riskier
"I don't have authority"	You have influence through information (Level 6)
"Let's just do what we can control"	Self-limiting your sphere of influence

Scope Boundaries

This command covers:

• Leverage point identification
• Higher-level alternative generation
• Prerequisite assessment
• Multi-level strategy design

Not covered:

• Full systems analysis (use /analyze-system)
• Detailed implementation planning
• Stakeholder communication