Google's Eric Schmidt on managing innovation through controlled chaos

Executive overview

Most managers try to eliminate chaos. At fast-moving companies, that instinct kills innovation. The best ideas don't come from the top — they emerge unpredictably from networks of smart people given freedom to experiment.

Schmidt's approach at Google combined three things: letting ideas flow freely bottom-up, making fast top-down decisions, and scaling winning ideas without hesitation.

Managed chaos — not managed order — is the engine of sustained innovation.

The case against control

• Companies that demand order suppress the lateral conversations where new ideas form.
• Humility is a prerequisite: leaders must accept they won't have the best ideas.
• Control freaks struggle because innovation can't be scheduled or predicted.
• Schmidt was convinced Google's auction-based ad pricing would bankrupt the company — it multiplied revenue.

How Google structured creative freedom

• Office layout mimicked Stanford grad school: four people per room, casual, communal food — deliberate signal of intellectual openness.
• 20% time let engineers devote one day a week to any project; it produced Gmail, Google Maps, Google News, AdSense.
• Hidden value of 20% time: employees could tell overbearing managers "I'll give you 100% of my 80%" — a structural check on management overreach.
• Product leaders could recruit any engineers they wanted, but only by convincing them the project was worth joining — no top-down assignment.
• Failure was expected and budgeted for; persistence means changing tactics, not repeating the same approach.

Decision-making as the counterbalance

• Free-flowing ideas only work when paired with fast, clear decisions.
• Google's weekly rhythm: staff meeting Monday, business review Wednesday, product decisions Friday — everyone knew when and where decisions landed.
• YouTube acquisition decided in 10 days; slow institutions "congest" and miss windows.
• Quick wrong decisions outperform slow right ones in almost every case.

Where ideas actually come from

• The heroic lone-inventor story is false: ideas improve through networks, not individuals.
• Person 89 might have the best idea — you can't predict the source in advance.
• Sharing problems company-wide is what enables unexpected solutions; siloed or fearful cultures block this.
• Mark Pincus distinction: instincts are right ~95% of the time, specific ideas only ~25% — kill ideas fast, protect the instinct.
• Google Earth revealed that cows align north-south due to magnetic resonance — discovered by researchers using Google Maps, an entirely unpredictable application.

Hiring for chaos

• Smart creatives — people who swap ideas and tackle challenges across functions — are the prerequisite for managed chaos.
• Schmidt and Larry Page reviewed all hiring to avoid "glue people": well-meaning generalists who add process but not value.
• Two strongest predictors of success in a knowledge economy: persistence and curiosity.
• Rocket scientists and Olympic athletes hired in technical and sales roles because early discipline signals capacity to operate under pressure.

Knowing when to scale

• Scaling too late costs compounding revenue; Europe became 50–60% of Google's profits because Schmidt moved immediately on international expansion.
• Schmidt sent Omid Kordestani to Europe with one instruction: don't come back without an office set up — within a week, London, Paris, and Hamburg leads were hired.
• Speed of scaling is itself a strategic decision, not an operational detail.