Use when you have a written implementation plan to execute in a separate session with review checkpoints
npx skills add defi-naly/skillbank --skill "good-strategy-bad-strategy"
Install specific skill from multi-skill repository
# Description
Richard Rumelt's framework for crafting and evaluating strategy through diagnosis, guiding policy, and coherent actions.
# SKILL.md
name: good-strategy-bad-strategy
description: "Richard Rumelt's framework for crafting and evaluating strategy through diagnosis, guiding policy, and coherent actions."
dimensions:
domain: [strategy, business, problem-solving, leadership]
phase: [strategy-evaluation, diagnosis, strategy-formulation, problem-solving]
problem_type: [strategic-clarity, bad-strategy-detection, challenge-diagnosis, action-coherence]
contexts:
- situation: "evaluating an existing strategy"
use_when: "testing for hallmarks of bad strategy (fluff, no diagnosis, goals as strategy)"
- situation: "facing a complex challenge"
use_when: "need to diagnose the crux before choosing a response"
- situation: "strategy feels like fluff"
use_when: "cutting through jargon to find substance"
- situation: "actions feel scattered"
use_when: "checking if actions are coherent and mutually reinforcing"
- situation: "goals stated without path"
use_when: "distinguishing real strategy from aspirations"
combines_with:
- playing-to-win # complementary strategy frameworks
- thinking-fast-and-slow # avoiding cognitive biases in diagnosis
- zero-to-one # identifying the crux of what to build
- hard-thing-about-hard-things # diagnosing crisis situations
contrast_with:
- skill: playing-to-win
distinction: "Rumelt focuses on KERNEL (diagnosis, policy, actions); Playing to Win on CASCADE of choices"
- skill: lean-startup
distinction: "Good Strategy is DIAGNOSIS and COHERENCE; Lean Startup is VALIDATION and ITERATION"
Good Strategy Bad Strategy Framework
The Kernel of Good Strategy
Good strategy has a basic underlying structure - the kernel - with three elements:
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β 1. DIAGNOSIS β
β What's going on? What's the challenge? β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ€
β 2. GUIDING POLICY β
β Overall approach to overcome the obstacle β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ€
β 3. COHERENT ACTIONS β
β Coordinated steps that carry out the policy β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
Strategy is NOT goals, ambition, or vision. Strategy is a coherent response to a challenge.
1. Diagnosis
The diagnosis simplifies reality by identifying the critical aspects of the situation. It defines the challenge.
Good diagnosis:
- Names the obstacle or challenge clearly
- Simplifies complex reality into actionable insight
- Often reframes the problem in a revealing way
Example: "Our problem isn't execution - it's that we're organized around products while customers buy solutions."
Questions:
- What's really going on here?
- What's the critical challenge we face?
- What pattern explains our situation?
2. Guiding Policy
The guiding policy is the overall approach chosen to cope with or overcome the obstacle. It channels action without fully defining it.
Good guiding policy:
- Creates advantage by anticipating actions of others
- Reduces complexity by ruling out many possible actions
- Provides direction while allowing adaptation
Example: "We will focus exclusively on the premium segment, refusing to compete on price."
The guiding policy is NOT a goal. "Be #1" is a goal. "Concentrate on high-margin niches where our service advantage matters" is a guiding policy.
3. Coherent Actions
Actions are coordinated and mutually reinforcing steps that implement the guiding policy.
Good coherent actions:
- Are coordinated - they work together
- Build on each other - not a random list
- Focus resources - concentrate rather than spread thin
- Are feasible - within organizational capability
Incoherent actions undermine each other or spread resources too thin.
The Four Hallmarks of Bad Strategy
1. Fluff
Superficial restatement of the obvious, using inflated words and concepts to create the illusion of high-level thinking.
Red flags: "Leverage synergies", "customer-centric transformation", "best-in-class solutions"
Test: Remove jargon. Is there any content left?
2. Failure to Face the Challenge
When the challenge isn't named, the strategy can't be evaluated or refined.
Red flags: Strategy documents that never state what problem they're solving.
Test: Can you state the challenge in one clear sentence?
3. Mistaking Goals for Strategy
Stating desired outcomes without addressing how to get there.
Bad: "Our strategy is to grow revenue 20%"
Good: "We'll grow by acquiring the #3 player and cross-selling our service contracts to their customer base"
Test: Does this explain how, not just what?
4. Bad Strategic Objectives
Objectives that are either too vague (fail to address critical issues) or too impossible (blue sky).
Bad objectives:
- "Improve customer satisfaction" (vague)
- "Become the global leader in 18 months" (impossible)
Good objectives are proximate - close enough to be feasible, concrete enough to be actionable.
Proximate Objectives
A proximate objective names a target that is achievable and will make a difference.
Far: "Solve climate change"
Proximate: "Reduce our facility emissions 40% in 3 years by switching to renewable power"
Characteristics:
- Close enough to be feasible
- Concrete enough to be evaluated
- Challenging enough to be meaningful
- Resolves ambiguity about what to do next
Sources of Power (Competitive Advantage)
Good strategy leverages sources of power - asymmetries that can be exploited:
| Source | Description |
|---|---|
| Leverage | Focusing effort on pivotal objectives |
| Proximate objectives | Achievable goals that create momentum |
| Chain-link systems | Integrated activities that are only as strong as the weakest link |
| Design | Tight integration where pieces fit together |
| Focus | Coordination of policies and actions |
| Growth | Competence-based expansion (not acquisition for its own sake) |
| Using dynamics | Riding waves of change (technology, regulation, demographics) |
| Inertia and entropy | Competitors' inability to respond |
Strategy as Problem-Solving
Rumelt's approach to crafting strategy:
- List the most important challenges facing the organization
- Prioritize - which is the crux?
- Diagnose - what's causing this?
- Generate options - what are possible guiding policies?
- Evaluate - which approach best leverages our strengths against the challenge?
- Define coherent actions - what specific steps carry out the policy?
Evaluation Checklist
Testing for Good Strategy:
- [ ] Diagnosis names a specific, critical challenge
- [ ] Guiding policy addresses the diagnosis directly
- [ ] Actions are coordinated and reinforce each other
- [ ] Resources are concentrated, not spread thin
- [ ] Objectives are proximate (feasible and concrete)
- [ ] The strategy creates or exploits an asymmetry
Red Flags for Bad Strategy:
- [ ] Heavy jargon with little substance (fluff)
- [ ] No clear statement of the challenge
- [ ] Goals stated without a path to achieve them
- [ ] Long list of priorities (everything is priority = nothing is)
- [ ] Actions that pull in different directions
- [ ] "Strategy" is just budgeting or planning relabeled
Quick Diagnostic Questions
When someone presents a "strategy," ask:
1. What is the challenge this addresses?
2. What is being ruled out by this approach?
3. Why will this work when alternatives won't?
4. What's the first coherent action?
5. How do the actions reinforce each other?
If they can't answer these clearly, it's probably not a strategy.
# Supported AI Coding Agents
This skill is compatible with the SKILL.md standard and works with all major AI coding agents:
Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.