Strategy as Choice
← Back
Roger L. Martin — Playing to Win
Strategy as Choice
Roger Martin's model for what strategy actually means, the Strategic Choice Cascade for winning, and why a plan is not a strategy.
Playing to Win (2013) — with A.G. Lafley
A New Way to Think (2022)
Revised definition (2025)
Martin's definition of strategy — original vs. revised
Original definition — c. 2000–2022
"An integrated set of choices that uniquely positions the firm in its industry so as to create sustainable advantage and superior value relative to the competition."
Used for over two decades. Five elements:
01Choices — to do some things and not others
02Integrated set — not a list; they fit together and reinforce each other
03Where to Play — a specific territory
04How to Win — a theory of competitive advantage
05Superior financial returns — the proof and reward
Revised definition — 2025
"An integrated set of choices that compels desired customer action."
Tightened to three essential elements. Where-to-Play and How-to-Win are still critical but now live inside the Strategy Choice Cascade rather than the definition itself. Superior financial returns dropped for simplicity — they're the outcome, not the definition.
01Choices — where the opposite is not stupid on its face
02Integrated set — choices across the cascade fit and reinforce one another
03Desired customer action — the thing the company does not control
The core distinction — strategy is not planning
Strategy
Uncomfortable, external, about winning
- An integrated set of choices that positions you to win on a playing field of your choice
- Specifies a competitive outcome you wish to achieve — it involves customers and competitors you don't control
- Must have a coherent, doable theory of why you will be better than competitors
- Cannot be proven in advance — requires accepting angst and uncertainty
- Should fit on a single page
vs.
Planning
Comfortable, internal, about resources
- A set of activities and projects the company says it's going to do
- Focused on resources you control — budgets, timelines, headcount, deliverables
- Does not require internal coherence — it's a list, not an integrated set
- Comfortable because you can prove each item is doable before committing
- Planning should follow strategy, never replace it — they are complements
The Strategic Choice Cascade — the five choices that make a strategy
The heart of Martin's model. Five interrelated questions, answered as an integrated set — not a fill-in-the-boxes exercise. The cause-and-effect logic between them is what makes it a strategy, not the answers alone.
01
Winning aspiration
What is our winning aspiration?
The purpose of the enterprise, its guiding mission. What does winning look like? Not just "doing well" — this is about defining a clear, inspiring competitive ambition that aligns all choices downstream. Goals and objectives live here.
02
Where to play
Where will we play?
The specific playing field you choose to compete on. Geographies, product categories, customer segments, distribution channels, vertical stages of production. Choosing where to play also means choosing where not to play — both are strategic choices.
03
How to win
How will we win where we've chosen to play?
Your theory of competitive advantage — the specific way you will be better than competitors in the chosen territory. Typically either a cost leadership or differentiation approach, but the key is having a coherent, doable logic for why customers will choose you.
04
Must-have capabilities
What capabilities must be in place?
The specific set of capabilities needed to execute the where-to-play and how-to-win choices. They must fit together and reinforce one another. Martin uses a "can't/won't test" — capabilities that competitors either cannot replicate or will not choose to replicate.
05
Enabling management systems
What management systems are required?
The systems, structures, and measures that build and maintain the must-have capabilities. Performance measurement, resource allocation, processes, and feedback loops that ensure the strategy is implemented and adapted over time.
Core principles of the model
01
Strategy is singular
One entity, one strategy. It is not a list of "strategies." It is a single integrated set of choices. If someone says "one of our strategies is…" they don't have a strategy — they have a list of initiatives.
02
Strategy is a creative act
Strategy is about designing something that doesn't yet exist. Conventional data-crunching and analysis can only tell you about the past — they are not well suited to creating an alternate future. Strategy requires imagination.
03
Choices must be made simultaneously
The five cascade questions are not answered sequentially — each decision links to and reinforces the others. They must be solved as matched sets where the whole is greater than the parts.
The planning trap
The comfort trap
Why organizations default to planning instead of strategy — and why it guarantees mediocrity.
The trap
Plans are comfortable because they involve things you control — budgets, timelines, resources. You can prove each line item is doable before committing. This feels safe and responsible.
The problem
Doing only things within your control won't add up to anything meaningful. You'll be busy executing a list of initiatives while competitors who made real strategic choices are winning the market.
The reality
Strategy involves specifying a competitive outcome that depends on customers and competitors you don't control. That uncertainty creates angst — and that angst is the signal you're doing strategy, not planning.
The escape
Accept that you cannot prove your strategy will work in advance. Lay out the logic — what would have to be true — and then watch, test, and tweak. That gives you the best chance of winning.
The Strategic Choice Structuring Process
"What would have to be true?"
Martin's signature question for making strategy choices. Instead of debating what's right, reverse-engineer the conditions each option needs to succeed — then test those conditions.
01Define the problem — a painful gap between desired state and current outcomes
02Generate possibilities — frame at least two genuinely different strategic options
03Ask "what would have to be true?" — for each option, identify the conditions about customers, competitors, capabilities, and costs that must hold for it to work
04Identify barriers — which conditions are you least confident about? Those are the barriers to choice
05Design tests — create specific, low-cost tests to see whether the least-confident conditions actually hold
06Choose and commit — based on test results, make the choice and build the integrated cascade around it
Bottom line
For Martin, strategy is not a plan, not a document, not a list of initiatives. It is an integrated set of choices that compels desired customer action. The Strategic Choice Cascade gives it structure. The "what would have to be true" question gives it rigor. And the willingness to accept uncertainty — rather than retreat to the comfort of planning — gives it power.
