Black Box Thinking

Why can’t we learn from failure? Because even in a no-blame culture it is human to deceive ourselves, and we don’t even know we’re doing it. It’s hard to change a belief we’re invested in, as Syed illustrates with true stories of miscarriages of justice, in which bad verdicts were maintained despite clear evidence they were wrong.

This book is recommended reading for anyone trying to learn, improve or innovate at work. In a well-researched book full of engrossing examples, Syed shows what a learning culture really looks like. He contrasts how aviation and healthcare respond when things go wrong. The black box of the title is the data capture system in planes, designed to reveal rather than obscure. With this and other data, aviation systematically uncovers, interrogates and, usually without blame, extracts the learning from accidents and near misses. This leads to changes in training and protocols which are rapidly disseminated worldwide.  By contrast, healthcare systems are mostly run as if medics are perfect and bad outcomes an unfortunate but inevitable occurrence, not to be dwelt on and certainly not publicised. Notable healthcare exceptions show how many thousands of patients could be spared the harm of medical errors every year, with a few simple changes.

The section on top-down and bottom-up learning has clear business applications. Examples from factory design, crime prevention and sport show that we need to value both. Putting theory into practice works in some situations but tinkering, experimenting and fine tuning can be just as effective. In business this bottom-up learning appears in the use of randomised controlled trials, supposedly introduced at Google by Marissa Mayer, and also part of the DNA of Capital One. The ultimate example perhaps is how a skinny Japanese student became a world champion speed-eater. There’s also a chapter specifically addressing how failure drives product innovation, featuring the familiar but still good story of Dyson’s invention of the bagless vacuum cleaner, which used both top-down and bottom-up learning to craft the final product, and to make Dyson’s fortune.