There are some kinds of failure we want to encourage. As I outlined in my last blog post – Should you “fail fast, fail often”? – intelligent failures are necessary if you want to innovate.
But not all failures are created equal. Most of the time, you want to avoid them if at all possible. So how do you do that?
To answer this question, I’m going to refer to Amy Edmondson’s excellent book Right Kind of Wrong: Why Learning to Fail Can Teach Us to Thrive.
What underlies basic failures?
‘Basic failures’ are caused by single, preventable errors. A simple mistake, a lapse in judgement, a slip-up… and the failure has happened. We’ve all done this many times before – and we’ll certainly do it again!
Though basic failures aren’t 100% preventable, their frequency can be reduced. You’ll need to start by examining what’s causing the basic failures in question – does it tend to be the same sort of thing? I’d suggest looking at the four general causes Edmondson outlines:
Inattention Not paying close enough attention can lead to errors, which can in turn lead to a basic failure. In a home setting this might look like getting distracted while a cake is in the oven (inattention), forgetting to take it out at the right time (error), and ending up with a charcoaly lump rather than a light and fluffy sponge.
Neglect Sticking to the baking example, neglect would be realising halfway through the recipe that you didn’t buy enough eggs. You meant to, of course, you just didn’t get around to it. In Edmondson’s words, “Neglect tends not to produce instant harm but rather allows the buildup that ultimately results in failure. Because we are forgetful and busy, it’s easy to put things off.”
Overconfidence If you’ve baked the same cake dozens of times and know the recipe back to front, that’s confidence. To assume that means you don’t need to consult recipes for other cakes is overconfidence. To quote Edmondson again, “simply not reflecting on the implications of a decision is a common underlying cause. People fail to draw on available information or even common sense. What was I thinking? is the vernacular”.
Faulty assumptions This is the “we’ve seen this before” or “we’ve always done it this way” problem. It might look like making the cake the same way you always have, because it’s always been a hit before… without stopping to check whether the person you’re making it for has an allergy.
How can you prevent basic failures?
As errors so often underlie basic failures, the first step is doing what you can to limit those errors. The first step here is removing the stigma – don’t blame people for their human fallibility! Befriend error. Befriend vulnerability. Show the people around you that you slip up all the time, and you’re neither scared by that nor by sharing that.
In any context where there’s a team involved, step two is focusing on safety – physical and psychological. It shows that your priorities are in the right place, and encourages people to speak up promptly when they see a mistake. Only by doing this can you catch errors in time to prevent them from leading to anything worse. And then you can learn from those mistakes – why did they happen? Do they point to anything systemic which needs to be dealt with?
Edmondson suggests a few systems to put in place at organisations to minimise basic failures:
Blameless reporting “The important thing to remember about errors is that they are unintended—and punishing them as a strategy for preventing failure will backfire. It encourages people not to admit errors, which ironically increases the likelihood of preventable basic failure.”
Preventative maintenance It’s just human nature to give more weight to what’s happening right now than to what may happen in future – ‘temporal discounting’. Edmondson recommends the 2016 In Praise of Maintenance episode of Freakonomics for a reminder of why we need to overcome this tendency.
Codification This is one of the ways you can systematise that maintenance, reducing its mental load. Simple checklists have revolutionised everything from aviation safety to frequency of infections in hospitals.
Required training Edmondson uses the example of Crew Resource Management (CRM), required aviation training first developed in 1979. By now it covers “leadership, communication, situation awareness, and hazardous attitudes. Many of CRM’s core tenets have been adopted by business”.
Failure-proofing This involves, as much as possible, thinking about how people think and putting up barriers to the likely mistakes they’ll make. It’s what underlies child-proof caps on medicine, auto-saving and many more everyday lifesavers (or dignity savers, like the ‘undo send’ button for emails first developed in Google Labs).
Can you reduce the likelihood of complex failures?
Complex failures have multiple causes, and untangling what led to them isn’t always easy. But at their root, you’ll often find simple errors and basic failures which weren’t caught early enough or taken seriously enough.
Because of this, the methods of basic failure prevention are also key to limiting complex failures. Focus on creating a culture of psychological safety, where errors and failures are reported quickly and blamelessly. Pay attention to those warnings, and welcome false alarms.
(Talking about use of the famous Andon Cords in Toyota factories, Edmondson explains that “a pulled Andon Cord that does not identify an actual error is framed as a useful drill […] a welcome education on how things go wrong and how to adjust so as to reduce that possibility.”)
By paying attention to the basics, leaders can create smart systems which accept errors as inevitable, but reduce their frequency. Which empower everyone to point out mistakes and prevent them growing. Which can differentiate between intelligent, experimental failure and the kinds of failure better avoided.