The Five-Question Cascade: Applying Playing to Win to a Product Line

Playing to Win says a product line's strategy is not a deck of goals — it's five nested choices (winning aspiration, where to play, how to win, capabilities, management systems) that have to logically reinforce each other. Pressure-testing a strategy means checking whether any one choice, if changed, would still support the four around it.

Quick answer: Roger Martin and A.G. Lafley's five choices — winning aspiration, where to play, how to win, capabilities, management systems — form a cascade, not a checklist. A strategy fails where two choices contradict each other, not where a single choice is "wrong" in isolation.

The Five Choices That Make a Real Strategy

A Playing to Win strategy is five specific, answerable choices arranged as a cascade: what you're trying to achieve, where you'll compete, how you'll win there, what capabilities that requires, and what systems keep it running. Each choice is both a hypothesis and a constraint on its neighbors — change one, and the others may no longer hold.

Martin and Lafley built the framework at Procter & Gamble and later distilled it in Playing to Win: How Strategy Really Works (Harvard Business Review Press, 2013). Their core complaint about most "strategy" documents still lands: they're financial targets and initiative lists wearing a strategy's clothes, with no chain of choices connecting the ambition to the work.

Richard Rumelt makes a related point in Good Strategy Bad Strategy (2011): a real strategy has a "kernel" — a diagnosis, a guiding policy, and coherent action — while a bad one is a restated goal with no logic connecting means to ends. Playing to Win is a more granular version of the same demand: five links instead of three, each testable against the others.

If you're mapping where this cascade sits relative to other strategy tools your team already uses, the product strategy field guide is a useful map of the broader landscape before you commit to one framework for a product line review.

Choice by Choice: What Each Link in the Cascade Actually Decides

Each of the five choices has a distinct job, and teams most often go wrong by solving the wrong one — spending a planning offsite refining the aspiration when the real gap is in where to play, or debating capabilities before how to win has been decided at all.

Winning Aspiration

The winning aspiration is a definition of competitive success specific enough to rule things out, not a mission statement. "Be the leader" is not an aspiration; "be the platform mid-market finance teams default to before they need a controller" is — it implies a market, an opponent, and a timeframe.

• It should name what winning means in competitive terms, not just "grow" or "delight customers."
• It should be stable across a planning cycle — if it changes every quarter, it was never doing the aspiration's job.
• It is the thing every other choice gets checked against, which is why leaving it vague is the most expensive mistake in the cascade.

Teams frequently conflate this with vision-statement writing, which is a related but distinct exercise — the guide to writing a vision statement people actually repeat is worth reading alongside this one, since a memorable vision and a competitively specific aspiration solve different problems.

Where to Play

Where to play is the set of choices about which customers, jobs, channels, geographies, and product categories you'll compete in — and, just as importantly, which you won't. It's a portfolio of bets, not a single market description, and it's the choice most often left implicit because naming exclusions feels like leaving revenue on the table.

Concretely, where-to-play choices cover:

1. Customer segment — which buyer and user profile, at what company size or life stage.
2. Product category — which job the product is positioned to do, and which adjacent jobs it deliberately ignores.
3. Channel — direct sales, self-serve, partner-led, marketplace.
4. Geography — which regions, and in what sequence.
5. Vertical — horizontal platform versus a specific industry's workflow.

Two tools sharpen this choice considerably. Naming the actual jobs customers are hiring your category to do — not just the demographic segment — is the core move in Jobs to Be Done thinking, and it turns a vague segment ("SMBs") into a specific arena ("teams replacing spreadsheet-based expense tracking"). Mapping the competitive and technical landscape those jobs sit inside — components, evolution, dependencies — is exactly what Wardley Mapping is built for, and it's a sharper where-to-play instrument than a market-sizing slide.

How to Win

How to win is the source of advantage inside the arena you just chose — the reason customers pick you over the alternative once they're already considering your category. Porter's framing in "What Is Strategy?" (Harvard Business Review, 1996) still holds: sustainable advantage comes from trade-offs, not from trying to be good at everything a competitor is good at.

In practice, how-to-win choices cluster into a few recognizable shapes:

How-to-win is frequently won or lost in the experience, not the feature list — which is where mapping the actual emotional arc of using the product, the kind of work covered in a customer journey mapping guide, earns its keep. A differentiation-based how-to-win that nobody feels in the actual journey isn't a real advantage yet.

Capabilities

Capabilities are the specific things the organization must be distinctively good at for the how-to-win choice to hold up — not a generic list of "strong engineering" and "great support," but the two or three things that, if you're not genuinely excellent at them, the how-to-win choice collapses.

• A cost-leadership how-to-win requires operational and pricing-engineering capability most competitors can't match.
• A differentiation how-to-win requires whatever produces the differentiated thing repeatedly, not once.
• An ecosystem lock-in how-to-win requires platform and integration capability that gets harder to copy the longer it compounds.

The test Martin proposes in his HBR writing on the framework is blunt: if you removed this capability, would the how-to-win choice still be credible? If yes, it's not actually a required capability — it's a nice-to-have wearing a strategic label.

Management Systems

Management systems are the mechanisms — planning cadences, metrics, incentives, hiring criteria, resourcing decisions — that keep the other four choices alive after the offsite ends. This is the choice most strategy documents skip entirely, which is exactly why so many good strategies never show up in daily work.

A cascade with four strong choices and no management-systems choice still fails — it just fails slowly, as ordinary operating decisions quietly drift back toward the old strategy.

This is the same gap covered in the strategy-deck-to-daily-execution problem: a strategy is only as real as the roadmap reviews, hiring rubrics, and OKRs that either enforce it or silently ignore it. Donald Sull and colleagues' research on strategy execution, published in Harvard Business Review, found a recurring pattern in large organizations: leaders below the top team frequently can't consistently name their own company's stated priorities — the choices existed on a slide, but no management system carried them downward.

A Filled-In Cascade: A Hypothetical Mid-Market Expense Product Line

Seeing the five choices filled in together, for one hypothetical product line, makes the reinforcement logic concrete in a way that abstract definitions don't. This example is illustrative only — a composite built to show how the choices interlock, not a real company's strategy.

Picture a product line inside a mid-market finance-software company: automated expense management for 100–1,000-employee companies that have outgrown spreadsheets but don't yet need a full ERP.

Read across the table and the logic should hold in both directions. The where-to-play choice (mid-market, outgrowing spreadsheets) is what makes the how-to-win choice (policy-aware automation over rigid rules) credible — an enterprise buyer with a mature controls function wouldn't value it the same way. The capabilities row exists because the how-to-win choice demands it, not because anomaly detection sounded impressive in a roadmap review.

Now stress-test it by breaking one link on purpose. If the where-to-play choice quietly expanded to include enterprise accounts — a common, unannounced drift — the how-to-win choice (automation over rigid rules) stops fitting a buyer who often wants rigid, auditable rules. The capability investment would be pointed at the wrong differentiator, and no amount of fixing management systems would repair that; the break is upstream.

Where Cascades Break, and the Reverse-Engineering Exercise That Catches It

Cascades break at the seams between choices, not inside any single choice — which is why reviewing each choice in isolation, the way most strategy templates are structured, misses most real failures. The fix is a directional test: for every choice, ask what has to be true one level up and one level down for this choice to still make sense.

Martin describes this as a logic-flow check — reading the cascade top to bottom asking "does this follow?" and then bottom to top asking "does this justify what's above it?" Both directions matter; a cascade can read fine going down and still fail going up if a capability was built for its own sake rather than because how-to-win demanded it.

A five-question reverse-engineering exercise

Run this on your own product line, or on a competitor's, using only what's publicly observable — pricing page, roadmap, job postings, marketing copy:

1. Infer where-to-play from the pricing page and case studies. Who is this obviously built for, regardless of what the "who we serve" copy claims?
2. Infer how-to-win from what's actually differentiated in the product, not the marketing adjectives — what would a switching customer say they got that they didn't have before?
3. Infer capabilities from job postings and integration partners. Hiring patterns reveal what an organization believes it needs to be good at more honestly than an "about us" page does.
4. Infer management systems from what gets measured publicly — investor updates, changelog cadence, support SLAs — as a proxy for what internal systems reward.
5. Check whether the inferred aspiration is consistent with all four. If it isn't, either the aspiration is stale, or one of the other choices has quietly drifted without anyone updating the story.

This exercise is as useful turned on your own product line as on a competitor's. It's often the fastest way to discover that the strategy everyone agreed to in a planning offsite and the strategy the roadmap actually implements have already diverged.

Bain & Company's long-running survey of management tools has repeatedly found strategic planning among the most-used tools by executives worldwide, while rating only middling on the satisfaction executives report with it — a gap consistent with cascades that exist on paper but were never checked for internal reinforcement.

Keeping the Cascade Intact as Decisions Multiply

A five-choice cascade agreed at a leadership offsite has to survive hundreds of smaller decisions made afterward — a pricing change, a roadmap trade-off, a new integration — each of which either reinforces the cascade or quietly erodes it. The hard part isn't writing the cascade once; it's giving every later decision something concrete to check itself against.

In practice this fails less because teams disagree with the strategy and more because the strategy becomes hard to find — it lives in a slide deck from a quarter or two ago, and the PM making today's prioritization call has no easy way to check a new feature against the where-to-play and how-to-win choices that were supposed to govern it.

Key Takeaways

• A strategy is five linked choices, not a document — winning aspiration, where to play, how to win, capabilities, and management systems, each constraining and constrained by the others.
• Cascades break at the seams, not inside a single choice — the most common failure is two choices that no longer logically support each other, not one choice that's individually wrong.
• Where to play is a portfolio of inclusions and exclusions. Naming what you won't do is as strategically load-bearing as naming what you will.
• How to win has to be felt in the actual product experience, not just claimed in positioning — differentiation nobody feels in the journey isn't a real advantage yet.
• Management systems are the most commonly skipped choice, and skipping it is why otherwise-good strategies quietly drift back toward old behavior.
• Reverse-engineering your own cascade from what's publicly observable — pricing, roadmap, hiring — is a fast, low-cost way to catch drift before a customer or competitor does.
• A worked, filled-in cascade with a deliberate stress test ("what breaks if this one choice shifts") reveals reinforcement gaps that reviewing choices individually never surfaces.