Cambrian Explosion

The Cambrian explosion is invoked as a model for understanding innovation: bursts of extraordinary variety produce not only new successes but also many failures. In the book’s account, Jeff Bezos uses this episode in evolutionary history to describe the Internet boom, reinforcing the argument that a productive system need not predict winners in advance if it can generate alternatives and recognize losers quickly.

The Analogy to the Internet Boom

Jeff Bezos, the CEO of Amazon, compares the Internet boom to the Cambrian explosion, described as the period in evolutionary history when more species were born and went extinct than in any other period. The comparison makes both abundance and loss essential parts of the same process: proliferation creates possibilities, but most do not endure.

The analogy therefore resists treating failure as evidence that experimentation itself has failed. If a period of exceptional innovation also produces exceptional extinction, then the losses accompany the generation of new possibilities rather than simply canceling them out.

Selection Without Foreknowledge

The book connects Bezos’s comparison to the familiar claim that governments cannot, and therefore should not try to, pick winners. It immediately qualifies that claim: no system seems especially good at identifying winners in advance. The difficulty is not confined to public decision-makers; companies themselves repeatedly commit substantial resources to products that fail.

Tens of thousands of new products appear each year, but only a small fraction succeed. The examples include steam-powered cars, picturephones, the Edsel, Betamax, and pen computing: distinct bets that companies made despite their eventual failure. Their variety gives the argument concrete force. The central question is not whether a system can avoid backing losers, but what it can do after a bet proves unsuccessful.

Failure as Part of a Productive System

The book’s answer is that successful systems must recognize losers and end them quickly. It sharpens the point by making generation part of the requirement: a system succeeds by producing many losers, identifying them, and killing them off. The Cambrian comparison thus frames innovation as a process whose fecundity and selectivity are inseparable.

That logic also explains why a tidy process may be less wise than a messy one. When no one can reliably select the eventual winners beforehand, a system that permits many experiments and can stop failures may outperform one organized around confidence in advance judgment. The Cambrian explosion is not presented as a detailed biological history; its role is to give this argument a vivid name and shape.