Summary of "How Innovation Works"

Core Idea

Innovation is recombination + iteration, not sudden genius—combine existing ideas, test variations rapidly, tolerate failure.
Affordability and reliability beat technical brilliance—practical adoption matters more than invention.
Innovation thrives in connected ecosystems—trade, dense cities, and diverse contacts spark breakthroughs; isolation kills capability.

Serendipity and accidents are primary—Post-its, Teflon, genetic fingerprinting emerged from solving different problems.
Teams beat solo genius—knowledge lives between heads; exchange and collaboration amplify results.
Decentralized systems outpace centralized ones—city-states and competitive markets innovate faster than unified empires.

Open beats closed—source 50% of ideas externally (Procter & Gamble model); don't rely solely on in-house R&D.
Use small, competitive teams—"two-pizza teams" (Amazon's model) reduce complacency and encourage experimentation.
Implement reverse-veto policy—allow ideas to bubble up even if most oppose them; one dissenting voice shouldn't kill novel concepts.
Fail fast and publicly—celebrate failures and share learnings transparently (Google X example).

Apply Amara's Law: Overestimate impact in 5–10 years; underestimate impact beyond 20 years—expect 15-year lag before predictions solidify.
Practice precedes theory—steam engines came before thermodynamics; breed first, understand genetics later.

Patent thickets block progress—strengthen IP only where essential (pharmaceuticals); don't create hedges covering single molecules.
Longer copyrights ≠ more creativity—music supply doubled post-Napster despite "piracy"; creators seek influence and fame.
Copyrights on scientific papers stifle innovation—taxpayer-funded research shouldn't be paywalled; embrace open-access.
Simple regulations enable entry; complex compliance favors incumbents.
Permissionless iteration is non-negotiable—regulatory approval for every variant wastes years and kills beneficial tech (Golden Rice, cellular phones delayed 39 years).

Expect predictable resistance playbook: Safety fears + vested interests + political paranoia emerge for every breakthrough (coffee, margarine, GMOs).
Precautionary principle backfires—higher safety standards for new tech vs. existing dangerous alternatives is backwards (copper sulfate still used by "organic" farmers).

Innovation increasingly means doing more with less—modern farming yields more food using less land, water, fertilizer.
Sustainable growth is achievable through efficiency, not depletion.
Inequality is about luxuries, not necessities—innovation makes basics universal; differentiation happens in premium markets.

Redesign R&D: Allocate 50% of innovation budget to external sourcing and partnerships; shrink internal labs.
Organize small competitive teams: Create "two-pizza units" with reverse-veto policies; celebrate transparent failures.
Reduce regulatory friction: Permit rapid iteration on variants; don't require full approval for each test cycle.
Open knowledge flows: Join or create trade networks, conferences, and dense collaboration spaces; isolate and you'll stagnate.
Expect 15+ years for impact: Plan for serendipity in decade-scale horizons; don't kill ideas failing in year 3.