Quote of the day by Andy Jassy: Invention requires two things: One, the ability to try a lot of experiments, and two, not having to...

Most companies say they want to innovate. What they actually want is to innovate without risk—to find the new thing, capture the upside, and pay nothing for the attempts that don't work out. That's not innovation. That's a wish.Andy Jassy has spent his career building systems that make the real version possible. As the architect of Amazon Web Services and now CEO of Amazon, he's watched more experiments fail than most executives ever greenlight. The Fire Phone was a public embarrassment. Amazon Destinations, Amazon Spark, and dozens of quieter bets simply disappeared. And yet Amazon is one of the most inventive companies on earth—not despite those failures, but in some sense because of them. Because the company was designed from the inside out to survive them.The quote below is the clearest distillation of that philosophy. It doesn't ask you to be bolder. It asks you to think structurally—about what invention actually requires, and what silently prevents it in most organisations."Invention requires two things: One, the ability to try a lot of experiments, and two, not having to live with the collateral damage of failed experiments."~ Andy Jassy, President and CEO, AmazonThis isn't a defence of recklessness. It's a precise two-part formula for how real innovation works—and most organisations only satisfy half of it.Most leaders understand that you need to experiment. What they underestimate is the second condition: the system around those experiments. If every failure carries catastrophic cost—to budgets, to careers, to reputations—then the rational move is to run fewer of them. You don't get more invention by telling people to be braver. You get it by redesigning the cost structure of being wrong.You don't get more invention by telling people to be braver. You get it by redesigning the cost structure of being wrong.Amazon's internal culture is built around this principle. The company has greenlit projects that looked absurd to outsiders—a physical bookstore, a voice-activated speaker, a cloud computing division spun out of what was supposed to be a retail business. Many bets failed publicly. Amazon Destinations. Amazon Spark. The Fire Phone. None of them derailed the company, because the infrastructure for failure—the willingness to shut things down cleanly, absorb the loss, and move—was already part of the operating model.AWS itself is the most powerful proof of the formula. It started as an internal experiment in reusable computing infrastructure. The cost of getting it wrong would have been contained. The upside of getting it right turned out to be enormous. You only discover that asymmetry if you're willing to run the experiment.The quote reframes what innovation actually requires. It's not just creative people with good ideas. It's a structural condition: the ability to be wrong cheaply. That's what the cloud, rapid prototyping, modular architecture, and small autonomous teams all have in common—they're mechanisms for reducing the collateral damage of failed experiments, not for guaranteeing success.Companies that invest heavily in planning before testing have the causality backwards. The point of experimentation is to generate information you couldn't have planned your way to. Jassy's quote is a reminder that the goal isn't to run better experiments—it's to run more of them, and to survive the ones that don't work out.Most people aren't running billion-dollar product lines. But the same logic applies at every scale. Launching a side project. Trying a new approach at work. Pitching an idea that might get rejected. The question Jassy is implicitly asking is: what's the actual cost if this fails? Often it's far lower than it feels. The collateral damage is mostly imagined. And once you reduce it—in your head or in reality—the number of experiments you're willing to run goes up dramatically.In an era where companies talk endlessly about innovation, most of them still punish failure. They run annual planning cycles that lock in budgets before any testing. They promote people who deliver, not people who learn. Jassy's two-part test is a useful diagnostic: does your organisation have the ability to run many experiments? And have you actually removed the collateral damage from the ones that fail? If the answer to either is no, the innovation agenda is mostly theatre.