Pattern-Breaking Inside a Big Company
Make small bets that can fail a lot — and hide them from the mother ship
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 4
- Confidence
- 90%
A method for pursuing breakthrough ideas inside an incumbent, where the corporate immune system reverts everything toward pattern-matching. The core move: treat the new thing as a separate, autonomous, low-visibility unit led by a maverick with real authority. Use it when a company wants genuine 0-to-1 innovation rather than an extension of its core business.
Origin
Maples argues everything about a pattern-breaking value-delivery system differs from a normal idea, so companies fail when they demand a new bet become a third of the business in a few years, or make it visible enough to become a career-limiting move. He points to Apple's iPhone, Lockheed's Skunk Works, and IBM's PC.
Core principles
- 01Don't make it too big to fail; make small bets that can fail a lot
- 02Visibility drags the project back into the tractor beam of acting like a pattern-matching corporation
- 03A coin that says 'can't lose' on one side says 'can't win big' on the other
How to run it
- 1
Separate it from the core organization
Create a unit organizationally distinct from the main business so it isn't reconciled against existing products.
Pro tip Lockheed's Skunk Works built a fighter plane fast precisely because it was a separate organization.
- 2
Give one maverick absolute power
Put someone with real authority and a maverick disposition in charge, as with Don Estridge on the IBM PC.
- 3
Keep it low-visibility in early innings
Deliberately make it not visible to the mother ship so it doesn't drift back toward corporate norms or become a career-limiting move.
Pro tip Consider a different time zone so would-be interferers are offline — one 0-to-1 PM did exactly this so his team could focus.
Watch out Making it visible invites the corporate immune system and turns failure into a career risk.
- 4
Fund it as asymmetric, fail-tolerant bets
Allocate a slice of profits (Vinod Khosla suggests ~10%) to projects likely to fail but with wildly asymmetric upside, judged on a different risk profile than the core.
Pro tip For M&A, first ask whether you're buying a pattern-breaking company to go in a radically new direction or an extension of your current business — the risk framing differs enormously.
In the wild
Lockheed built a fighter plane in a very short time using an organization totally separate from the main one, headed by someone with absolute power, kept separate and low-visibility.
→ Cited as the archetype of autonomous, maverick-led corporate pattern-breaking.
The iPhone was a whole new thing that couldn't be reconciled with the Mac or even the iPod business, so it had to be treated as a totally separate thing, driven by founder Steve Jobs.
→ A pattern-breaking product succeeding by not being forced to fit the existing business.
Common mistakes
Requiring the new bet to be a third of the business in three years
Demanding scale on a corporate timeline is a bad strategy that kills breakthrough bets before they can develop.
Making the project too visible
Visibility pulls it back into pattern-matching behavior and turns failure into a career-limiting move.
Is it for you?
Best for
Corporate leaders and intrapreneurs pursuing genuine 0-to-1 innovation inside an incumbent
Not ideal for
Incremental improvements or extensions that belong inside the core business
From the transcript
“I need to make small bets that can fail a lot”
“when you make it visible you start to drift it back into the tractor beam of acting like a pattern matching Corporation”
“any coin that says can't lose bad on one side of it might as well say can't win big on the other side”
From the episode
Pattern Breakers: How to find a breakthrough startup idea
Mike Maples, Jr. (Founding Partner at Floodgate, ex-Product