Roger Martin vs Richard Rumelt: Strategy as Choices or Strategy as a Kernel?
Comparing Roger Martin's Playing to Win cascade of choices with Richard Rumelt's kernel of diagnosis, guiding policy, and coherent action: how each defines strategy and how to use both.
Two of the most-used modern definitions of strategy come from Roger Martin, whose Playing to Win cascade became a standard workshop tool, and Richard Rumelt, whose kernel of good strategy is a standard test. People compare them because both rescue strategy from vague planning, and because they offer different structures for doing so.
What is the core difference?
Martin asks what choices a firm must make and in what order. Aspiration, where to play, how to win, capabilities, and management systems form a cascade in which each answer shapes the next. The framework gives a team a structure for the conversation.
Rumelt asks whether a plan deals with a specific challenge. The kernel is a test for whether strategy exists at all: is there a diagnosis, a policy that responds to it, and actions that fit together?
One is a map of the decisions. The other is a check on the thinking behind them.
What does Martin say?
Roger Martin, a Canadian strategy writer and former dean of the Rotman School of Management, wrote Playing to Win (2013) with A.G. Lafley, then chief executive of Procter & Gamble. The book's five-question cascade is drawn from their work together at P&G.
Martin also wrote The Opposable Mind (2007), on integrative thinking, and Fixing the Game (2011). A recurring theme is that strategy is a choice about how to win, which means choosing against alternatives and accepting that you will not do everything. His writing is a practitioner's and a scholar's blend, with company examples.
What does Rumelt say?
Richard Rumelt, a professor emeritus at UCLA's Anderson School, wrote Good Strategy/Bad Strategy (2011). He names the features of bad strategy as fluff, avoiding the challenge, mistaking goals for strategy, and bad objectives. His kernel of good strategy has a diagnosis, a guiding policy, and coherent actions.
He also emphasizes focus, concentrating resources on a pivotal objective. The book relies on cases and his consulting experience, and he writes with sharp critiques of common corporate practice.
How do they compare?
| Dimension | Martin | Rumelt |
|---|---|---|
| Form | A cascade of five linked choices | A kernel of three elements used as a test |
| Starting point | Winning aspiration | Diagnosis of the challenge |
| Use | A structured process for a team to work through | A check on whether a plan counts as strategy |
| Core warning | Strategy is choosing, not listing | Bad strategy hides in goals, fluff, and templates |
| Co-author or source | With A.G. Lafley, from P&G experience | Rumelt's research and consulting |
| Risk when misused | Filling boxes without real choices | Diagnosing without turning it into choices |
| Best known for | Playing to Win (2013) | Good Strategy/Bad Strategy (2011) |
When does each one fit?
Martin's cascade fits when a team needs a structure to talk through a strategy and wants to see how choices connect. It suits workshops and planning cycles where people need a shared vocabulary.
Rumelt's kernel fits when a plan already exists and you want to know whether it is strategic or simply a list. It suits reviews, board discussions, and moments when a document feels hollow.
When the challenge is unclear, start with Rumelt's diagnosis. When the challenge is clear and the question is what to commit to, Martin's cascade gives the sequence.
What does this look like in practice?
A regional food company must decide how to respond to cheaper rivals. A Rumelt-style start writes the diagnosis: customers buy on price in the main channel, where larger rivals have lower costs, while a growing group of restaurants values freshness and traceability. That challenge rules out a price war.
A Martin-style cascade then fills the choices. Aspiration: be the preferred supplier for quality-led restaurants in the region. Where to play: restaurants, not supermarkets. How to win: freshness and traceability. Capabilities and systems follow. The diagnosis supplies the reason, and the cascade supplies the structure.
"My business is [description] and the main strategic problem seems to be [problem]. First, write three candidate diagnoses in one sentence each, following Rumelt, and say what each would rule out. Then take the most convincing one and fill in Martin's five questions: winning aspiration, where to play, how to win, required capabilities, and management systems. Flag any answer that does not follow from the diagnosis."
Why it works: the diagnosis keeps the cascade from becoming a form-filling exercise.
Can you use both together?
Yes, and the combination answers the usual weakness of each. A cascade without a diagnosis can contain reasonable choices that address nothing in particular. A diagnosis without a structure can end in a clever insight and no clear commitments.
Both ask you to choose and to give up options. If your output keeps everything, neither has been applied.
"Here is our current strategic plan: [paste]. Apply Rumelt's test: where is the diagnosis, the guiding policy, and the coherent actions? Then map it onto Martin's cascade and tell me which of the five choices are missing or unstated, and which choices contradict each other."
Why it works: each framework exposes gaps the other can miss.
For a smaller company facing a bigger rival, see Competing Against a Much Bigger Rival Without Copying Them.
Where to go next
Roger Martin and Richard Rumelt both sit in the Business Strategist category. For the originals, read Playing to Win by Martin and A.G. Lafley, and Good Strategy/Bad Strategy by Rumelt. To work through a strategic choice, Decision Brief is a $79 tool built for it.
Frequently asked questions
How do Martin and Rumelt define strategy differently?
Martin, in Playing to Win (2013, with A.G. Lafley), defines strategy as an integrated set of choices that positions a firm to win, organized as a cascade from aspiration to management systems. Rumelt, in Good Strategy/Bad Strategy (2011), defines good strategy by its kernel: a diagnosis, a guiding policy, and coherent actions. Both are about choice, and they organize it differently. Reading both authors' own explanations of the term shows how much of the difference is emphasis and how much is substance.
What is the Playing to Win cascade?
It is a sequence of five linked questions: what is our winning aspiration, where will we play, how will we win, what capabilities must be in place, and what management systems are required. Each choice constrains the next. Martin and Lafley developed it from their work at Procter & Gamble, and it is presented as a framework a team can fill in. Each answer should constrain the next, so if a later choice would work equally well under a different aspiration, the cascade is not yet doing its job.
Does Rumelt reject frameworks like the cascade?
He is wary of frameworks used as templates for filling in boxes, and argues that bad strategy often hides in formulaic documents. His kernel is a test rather than a template. A cascade can pass or fail the test depending on whether its choices follow from a real diagnosis. The two are not in conflict if the cascade is used thoughtfully. Rumelt's own test is whether the choices respond to something real, which is the question a cascade alone does not ask.
What is integrative thinking?
It is Martin's term, set out in The Opposable Mind (2007), for the ability to hold two opposing ideas in mind and produce a new answer that contains elements of both, instead of choosing one. He argues that successful leaders tend to do this. It is a way of approaching strategic choices, separate from the cascade. Martin argues that integrative thinkers resist forced either-or choices and look for a better option, which can help when deciding where to play and how to win.
Which is easier for a team to use?
The cascade gives a visible sequence a team can work through in a workshop. The kernel is a smaller test that can be applied to any plan quickly. Teams that want a structured process often prefer the former, and those who want a sharp check on an existing plan often prefer the latter. This is an observation about use, not a documented ranking. Whichever you start with, working through one real decision teaches more than reading about either framework.
Can AI help me with either?
It can draft answers to the five cascade questions, test whether a plan has a diagnosis, and propose alternatives to challenge your choices. It cannot judge which choice your organization can execute. Use it to widen options and expose weak spots, and then test the result with people who know the business. Give it the full context, ask it to argue against each choice, and keep the final decision with the people who must carry it out.
Written by Gareth Hoyle. Last updated 8 October 2026. Part of the authority.md guides library.
More guides.
Daron Acemoglu vs Ha-Joon Chang: Do Good Institutions Cause Growth, or Follow It?
Comparing Daron Acemoglu's institutions-first account of national prosperity in Why Nations Fail with Ha-Joon Chang's historical critique of the institutions literature: two views of development, with the main point of disagreement. General education.
Bill Ackman vs Carl Icahn: Two Styles of Activist Investing. Not Investment Advice
Comparing Bill Ackman's public, thesis-driven activism with Carl Icahn's decades of corporate pressure campaigns: how each approaches companies, boards, and the public. Not investment advice.
Adam Smith vs Ha-Joon Chang: The Founder of Market Economics and One of Its Historical Critics
Comparing Adam Smith's Wealth of Nations and Theory of Moral Sentiments with Ha-Joon Chang's history-based critique in Kicking Away the Ladder: what Smith actually argued, what Chang adds, and where they differ. General education.