Good Strategy Bad Strategy by Richard Rumelt — Book Notes

Author: Richard Rumelt (2011)

The book’s premise: most things called “strategy” aren’t. They’re ambitions, financial targets, vision statements, or to-do lists. Real strategy is a coherent response to a specific challenge — and it’s rare because it requires making choices that exclude things.

Bad Strategy

Rumelt’s four hallmarks of bad strategy:

1. Fluff: jargon inflated to sound profound. “Our fundamental strategy is one of customer-centric intermediation.”
2. Failure to face the challenge: if you haven’t defined the problem, you can’t evaluate the strategy, and you can’t improve it.
3. Mistaking goals for strategy: “grow revenue 20%” is an ambition, not a plan. Strategy is how.
4. Bad strategic objectives: either a “dog’s dinner” (a long list of everything, prioritizing nothing) or “blue sky” (restating the desired outcome without addressing how to get past the obstacle).

Bad strategy isn’t the absence of good strategy — it’s an active choice to avoid the hard work of analysis and the pain of choosing.

The Kernel of Good Strategy

Every good strategy has three parts:

1. Diagnosis: a theory of what’s going on: naming the critical challenge, simplifying the overwhelming complexity of reality to the aspects that matter most. Often the diagnosis alone is transformative (Kmart vs. Walmart: the diagnosis that Walmart’s advantage was the network, not the store).

2. Guiding policy: the overall approach chosen to overcome the obstacles in the diagnosis. Not a goal — a method. It rules things out. It channels action in certain directions without specifying every step.

3. Coherent action: coordinated steps that implement the guiding policy. The actions must be consistent with each other; resource allocations, policies, and maneuvers that don’t work together (or work against each other) are the signature of no-strategy.

Missing any of the three → not a strategy.

Sources of Power

The middle of the book catalogs where strategic advantage actually comes from:

• Leverage: concentrating effort at a pivotal point (“the point where a small adjustment releases a disproportionate result”).
• Proximate objectives: targets close enough to actually hit. Kennedy’s moon shot worked because it was feasible: each sub-problem was already understood to be solvable.
• Chain-link systems: when quality is limited by the weakest link, incremental improvement anywhere else is wasted. (Also the moat: excellence across a chain is hard to copy — IKEA.)
• Design: strategy as designed configuration of resources and actions, not decision. The tighter the fit between the pieces, the stronger and more fragile the advantage.
• Focus, growth, advantage, dynamics, inertia, entropy: each a lens on where power comes from or leaks out.

Key Takeaways

• The most useful question when handed a “strategy”: what is the diagnosis? If no one can state the challenge, it’s decoration.
• Strategy is as much about what you won’t do. If it excludes nothing, it’s a wish.
• Prefer proximate objectives — strategy that requires a miracle at step three isn’t strategy.
• New ideas rarely win by being better everywhere; they win by being decisively better at a pivotal point (leverage).
• “The kernel” is a portable tool: it works for a company, a product launch, a career move, or a country.

Why It Stuck With Me

The kernel is a 30-second test that most strategy documents fail. Diagnosis → guiding policy → coherent action forces the sequence most planning skips: understand the problem before choosing the approach, choose the approach before listing the actions.