A framework by Richard Rumelt
Richard Rumelt says every good strategy has the same three parts. A diagnosis that names the challenge. A guiding policy that says how you will deal with it. Coherent action that carries the policy out. He calls the three together the kernel. Drop any one of them and what you have left is a goal.
What the word priority means
It doesn't mean the grab bag of everything that you can think of that might matter. It means what's first.
One client arrived with 17 priorities. Rumelt calls that the opposite of a policy.
The framework
The diagnosis answers one question: what is going on here. Not the symptom, the challenge. Rumelt spends most of his time here, because this is the part organizations skip. Skip it and everything downstream goes wrong without anyone noticing.
The guiding policy comes next. It is the approach you are taking, and Rumelt calls it the core of a strategy. It says how you will deal with the situation you just named. The word most people reach for here is priority, and they use it wrong.
it doesn't mean the grab bag of everything that you can think of that might matter, it means what's first
He gets the image from air traffic control. A tower gives a pilot priority over the next three planes on the runway. It means first, in a queue, right now. A list of 17 things is not a queue. Rumelt had a client bring him that list, and he calls it the opposite of a policy. It held everything the company hoped would happen next year. Market share in China. Lower costs. Better safety. All at once.
Then comes the part he says is the easiest to leave out, because people like to think of strategy as a high-level conceptual thing. Coherent action. The actual moves, and they have to point the same way. A guiding policy with no action attached is a sentiment, and actions that fight each other cancel out.
Practice
Run it as a test on the document you already have, then rebuild the parts that fail.
Name the challenge in a sentence. Separate the symptom from the cause. If your strategy document opens with a target instead of a problem, you have not started yet.
Rumelt borrows the word from climbing: the hardest section of the route that you can still get past. List every challenge, then judge each on importance and on whether it can be solved. Work the one that scores on both.
Say how you are dealing with the crux, and say what you are not doing. Rumelt's answer on how many things to hold at once is blunt: a few. We work best when we concentrate.
Write the moves the policy implies. Then read them as a set and look for pairs that pull against each other. Coherent means they compound, not that each one looks sensible alone.
Rumelt lists nine, including leverage, focus, chain-link systems and inertia. You need a reason the strategy will work. If you cannot say which advantage you are applying, you are hoping.
A choice challenge has known options and unclear trade-offs. An engineering-design challenge can be modelled and tested. A gnarly one has neither. The type decides the method, and treating a gnarly problem as a choice is how teams end up voting on guesses.
Boundary conditions
Rumelt's answer is that bad strategy is not an absence. It is a thing with its own shape, and he can list the hallmarks. The first one is easy to spot once you have the word for it.
bad strategy also is fluff, people will use fancy words to describe their situation
He notes that the term word salad has caught on since he wrote the book, and that plenty of strategy writing qualifies. But fluff is the showy failure, not the common one. Asked to rank them, he puts something duller first.
lacking diagnosis is one of the key reasons for bad strategy
Fluff and incoherent actions are fun to describe. He says they are also less common than you would think. The missing diagnosis is the real one, followed by mistaking goals for strategy. At the LSE he showed what the gap looks like from outside. He walked through flatlined growth, the European debt crisis and a school district losing 40% of its kids.
and you will find over and over again that when you enter these arenas that people are putting forth ideas about what to do with no diagnosis
The pattern under all four is the same. A diagnosis commits you, and a list does not. That is why the list is more popular.
For the flip side of concentrating force against a rival, read Peter Thiel's contrarian question. It argues the better move is to avoid the fight.
Primary sources
Two recordings, both with Rumelt teaching rather than being summarized. The Lenny episode runs the kernel for an operator audience. The LSE lecture applies it to public problems, which is where the missing diagnosis is easiest to see.
Richard Rumelt · Lenny's Podcast
What Priority Actually Means
"it doesn't mean the grab bag of everything that you can think of that might matter"
Richard Rumelt · Lenny's Podcast
Fluff, and the Word Salad Problem
"bad strategy also is fluff, people will use fancy words to describe their situation"
Richard Rumelt · Lenny's Podcast
Which Failure Mode Is Most Common
"lacking diagnosis is one of the key reasons for bad strategy"
Richard Rumelt · LSE Events
Action Proposed Without a Diagnosis
"people are putting forth ideas about what to do with no diagnosis"
Where experts disagree
Richard Rumelt
assumes you will have rivals and tells you how to beat them. Diagnose the crux, concentrate force at the decisive point, and apply a named source of power such as leverage or focus. Competition is the terrain, not a mistake.
Peter Thiel
calls competition a loser's game. Profits get bid away in a crowded market and companies converge on each other, so the real move is to build something new enough that nobody is standing there to fight.
Which one applies depends on whether an unclaimed position exists. Thiel is right when you can find one, and Rumelt is right the rest of the time, which is most of the time. Both agree on the failure they are arguing against: spreading effort thinly across a market you never diagnosed.
Want the full playbook?
From Richard Rumelt, Annie Duke, Charlie Munger, and 8 more. Drop one .md into Claude, Cursor, or ChatGPT. Your AI cites practitioners, not guesses.
Instant .md download · One-time purchase · No subscription
Related frameworks
Strategy
Strategic Inflection Points
Andy Grove
When one force grows 10x, the industry changes shape. Grove says the frontline sees it before the CEO does.
Strategy
The Four Forms of Leverage
Naval Ravikant
Labor, capital, code and media. The first two need someone's permission. The last two are why the newly rich got rich.
Distribution
The Atomic Network
Andrew Chen
The smallest group that has to be connected at the same time, doing the same thing, before the product is worth using. Two or three for Zoom, hundreds for Tinder.
Decision Making
The Flywheel Effect
Jim Collins
Breakthrough is cumulative, not decisive. Asking which push caused it is, in Collins's words, a nonsense question.
