StrategyAdvanced7 min read

Discovery Driven Planning

Discovery Driven Planning (DDP), developed by Rita McGrath and Ian MacMillan in their 1995 HBR article, is a planning method for ventures where most of what you 'know' is actually an assumption. Instead of starting with revenue projections (which everyone fakes), you start with a Reverse Income Statement: How much profit must this business produce to be worth doing? What revenue does that imply? What units must we sell to hit that revenue? What customer behavior must hold for those unit sales? Then you make every assumption explicit on an Assumption Checklist, attach a checkpoint where each assumption gets tested with real data, and convert assumptions to knowledge before spending the next dollar. The discipline: you fund the next milestone, not the whole plan.

Also known asDDPReverse Income StatementAssumption-to-Knowledge PlanningMcGrath Method

Challenge a friend Browse library

The Trap

The trap is treating the spreadsheet as a plan rather than a list of bets. Most 'business cases' for new ventures are reverse-engineered: someone wants to do the project, they back into the numbers that make it look acceptable, and the assumptions that make those numbers possible are buried in cell comments nobody reads. DDP makes the assumptions the deliverable. If you can't list 30+ specific, testable assumptions behind your $50M revenue forecast — and rank them by how badly you'd be wrong if they're wrong — you don't have a plan, you have a wish.

What to Do

Run the DDP loop on any new venture before approving full funding: (1) Write the Reverse Income Statement — required profit, implied revenue, implied unit sales. (2) Build the operations specs — what must the business actually do (unit cost, conversion rate, sales cycle, churn) to produce those numbers? (3) List every assumption embedded in the specs. Rank by impact-if-wrong. (4) Define checkpoints — the cheapest test that converts each top assumption to knowledge. (5) Fund only the next checkpoint, not the whole plan. (6) After each checkpoint, redo the spreadsheet with new data and re-rank assumptions. If the Reverse Income Statement no longer holds, kill the venture or pivot — don't 'adjust the assumptions' to keep it alive.

Formula

Required Revenue = Required Profit ÷ Required Margin → Required Unit Sales = Required Revenue ÷ Avg Selling Price

In Practice

Hypothetical: A consumer-electronics incumbent uses DDP to evaluate a smart-home subscription service. Required profit: $40M/yr. Required margin: 20%. Implied revenue: $200M. ASP: $15/mo, so implied subscriber count: 1.1M. Their assumption checklist surfaces 47 assumptions; the top three by impact-if-wrong are (a) blended CAC < $80, (b) annual churn < 18%, (c) attach rate to existing hardware buyers > 9%. Instead of funding the full $90M build, they fund a $4M, 90-day Florida-only pilot designed specifically to test those three numbers. Real CAC comes back at $190, churn at 31%. The Reverse Income Statement now requires 3.4M subscribers — implausible. They kill the project at $4M instead of $90M. The point of DDP isn't optimism or pessimism — it's killing bad ventures before they kill you.

Pro Tips

01
The Reverse Income Statement should make you uncomfortable. If the numbers required look easy, you've either picked an unambitious venture or you're underestimating costs. The whole point is to surface the heroic assumptions early.
02
Rank assumptions by impact-if-wrong, not by likelihood-of-being-wrong. The assumption you're most confident about can still be the one that destroys the business if it's wrong — confidence is not the same as evidence.
03
Write the kill criteria before the checkpoint, not after. After you've spent $4M, the team will rationalize any data into 'directionally positive.' Pre-commit to the threshold that triggers a kill.

Myth vs Reality

Myth

“DDP is just 'lean startup' for big companies.”

Reality

DDP predates Lean Startup by 14 years and operates at a different altitude. Lean Startup focuses on product-market fit at the experiment level. DDP focuses on financial viability at the venture level — it asks not 'do customers want this?' but 'even if they do, can the unit economics ever produce the required profit?'

Myth

“DDP slows down innovation by demanding too much rigor up front.”

Reality

DDP accelerates innovation by killing bad ventures faster. The slow death is the venture that limps along for three years on the strength of unfalsifiable optimism. DDP's checkpoint discipline is what lets you reallocate capital from losers to winners on a 90-day cycle instead of a 3-year cycle.

Try it

Run the numbers.

Pressure-test the concept against your own knowledge — answer the challenge or try the live scenario.

🧪

Knowledge Check

Challenge coming soon for this concept.

Decision scenario

The Forced Pivot

You're VP of New Ventures at a $2B industrial company. Eighteen months ago, you funded a $30M smart-equipment-as-a-service venture. Reverse Income Statement said: $25M revenue Y3, $5M profit, 1,250 customers at $20K ACV. After 18 months, you have 180 customers at $14K real ACV, churn at 24% (vs. assumed 12%), and $19M of the $30M spent. The team's plan: 'one more big enterprise deal will get us back on track.'

Capital Spent

$19M of $30M

Customers (Plan vs Actual)

Plan: 1,250 / Actual: 180

ACV (Plan vs Actual)

Plan: $20K / Actual: $14K

Annual Churn (Plan vs Actual)

Plan: 12% / Actual: 24%

Decision 1

You redo the Reverse Income Statement with real numbers. To still hit $5M profit at the new ACV, churn, and operating cost reality, you'd need ~3,500 net customers — nearly 20x today's count and 3x the original plan. The assumption checklist shows three foundational assumptions are now falsified, not just 'directionally off.' The team's 'one big deal' framing is hope, not a plan.

Approve the remaining $11M and another $5M bridge to give the team 'a fair shot' at the enterprise pipeline. The original business case was approved by the board and a kill decision is politically painful.Reveal

Twelve months later: $36M total spent, 240 customers, churn unchanged, no whale deal closed. You write off the venture in a board meeting where the same directors who approved it ask why no one raised a flag earlier. You did, but not loud enough.

Total Spend: $19M → $36MCustomers: 180 → 240Eventual Outcome: Same kill — $17M later

Trigger the pre-committed kill criteria. Stop the spend now, redeploy the remaining $11M to the two ventures in your portfolio that are beating their own Reverse Income Statements, and run a public post-mortem on which assumptions were wrong and why nobody escalated sooner.Reveal

Painful in the short term — the team is upset, the original sponsor loses face. But $11M redeployed funds two ventures that are tracking ahead of plan, one of which becomes a $40M revenue line within 24 months. The post-mortem changes how the company reviews ventures: every quarterly business review now starts with 'which top assumptions have flipped?' DDP discipline is finally cultural, not just procedural.

Capital Saved & Redeployed: $11M to higher-EV venturesPortfolio Net Outcome: +$40M revenue from redeploymentCultural Change: Assumption discipline institutionalized

Related concepts