How Amazon Deals With Failure

In your experience at Amazon, were there any practices that struck you as really effective in learning from failure?

One is that the company openly talks about it. That’s the starting point. The worst thing is being ashamed by failure. You failed as well. That shouldn’t be a big discovery. Again, it’s that recognition — “I messed up, I made a mistake, and here’s what I’ve learned as a result. Here’s what I could do better next time.”

It’s creating that open dialogue and awareness around it, but you can only do that if you feel safe to share. “I did fail; I made a mistake here, I made a mistake there, and I’ve learned from it.” That’s growth and development. Dismissing and hiding it, nobody can learn from that.

That’s at the forefront of how Amazon specifically did very well. “Things went wrong.” And asking the question, “What could we do better?” Having spent some time in the operations business, as soon as we finished Q4, we’d say, “Let’s go back and do a post-mortem — what happened? What went well, what didn’t work?” You focus on the stuff that went well, to say, “Let’s do that again.”

But very quickly, you move on to, “What didn’t go well? What can we do to improve it?” Eight months later, when you’re getting ready for the next big period, you dust off that document and say, “Are we ready? Let’s start with the things we learned last time. What went well? Let’s make sure we do those again. What went badly? Have we done enough to avoid those things happening again?”

You build that into the organisational DNA. There’s nothing embarrassing or shameful around it. It helps the organisation grow, no different to an individual. If you ask me for feedback and I give you feedback, it’s great, you’ve got the feedback, and you can use it to try and improve.