Amy Edmondson: “Most of us need a playbook for failure”

In her latest book "The Right Kind of Wrong", named Financial Times Business Book of the Year 2023, Professor Amy Edmondson invites us to reconsider the role of failure—in business, in the classroom, and in our private lives.

On the evening before she addressed a full house at Newton University, Edmondson sat down with Newton Today to share her insights on mastering the science of productive failure. Her approach doesn't discard the Silicon Valley mantra of "fail fast, fail often" when it fits. Yet, she believes the majority of us need a playbook to learn how to strike the delicate balance between embracing necessary risks and recognising when to adhere to tried-and-true methods. Read on for the full conversation.

We all know about failure. A falsifiable hypothesis is a key part of the scientific method, and Silicon Valley has embraced “fail fast, fail often” as a mantra. I know you draw a distinction between your approach and that attitude. How would you characterise that distinction?

Well, I'm perfectly comfortable with the idea of “fail fast, fail” often under certain conditions. The distinction is that the silicon valley logic applies well to certain kinds of situations and doesn’t work at all in other kinds of situations. So, my idea is that we need to be more thoughtful about context and more precise and make important distinctions that allow us to be effective.

So, under what conditions does the advice fail fast, fail often make sense? Well, very simply, in new territory where there is no playbook, there is no existing knowledge for how to get the result we're trying to get.

And let's just illustrate that one. If you are starting a brand-new company, you have no choice but to do the very best you can in inherently new territory. And you may succeed, you may fail, and you must appreciate that at the outset.

Two, it must be in pursuit of a goal, and three, with a hypothesis — a hypothesis drawn from doing your homework, meaning you've taken the time to find out what  we already know, what we do not know, and you have good reason to believe that your startup or experiment in the lab might work.

And then fourth, keep it small. If you're an innovator and you have something you want to test, don't invest all of your resources in it. Don't put all your eggs in the same basket.

But if you're just trying to cook a batch of cookies, use a recipe. Why have a failure?

Is that a level of nuance that you see being understood well in startups and by entrepreneurs?

I think if you stopped them in action and said, “let's talk about this”, they would say, “of course, what you just said is obvious”.

But I guess my point is that we don't have to worry about entrepreneurs. They're out there doing their thing. We don't have to worry about scientists. They're doing their thing. It's the rest of us who may not be taking the smart risks that we can take to have better results in our teams at work or in our everyday lives.

And the rhetoric, “fail fast, fail often, it's the secret to life” —it doesn't help us. Because at a deep level, we know it’s not right. Failure is not good — unless it's failure in new territory, with a hypothesis, in pursuit of a goal, and the failure is not very big.

And so I just think people need a playbook. The natural innovators, many of the people I write about in my book, ‘Right Kind of Wrong’, celebrity chefs, elite athletes, scientists, they didn't need to read my book to do their thing. They were already doing it, and it has a fundamental logic to it. But the rest of us do need a playbook.

Your book covers all kinds of failures—from academic failures that lead to world-changing breakthroughs, to forgetting to charge your phone and being late for a meeting. Do these failures have enough in common that we can develop a playbook to tackle them all?

It's a very good question. I think the answer is yes. And it's maybe because I think conceptually. I like ideas, and so I think in terms of categories about what makes something like something else, even if it's very different.

And I think there are enough very good and consistent answers to that question to say “yes”. It's a helpful framework, even though it’s almost absurd that it applies to such different phenomena.

Cultural attitudes to failure can vary significantly. In your research, have you found certain cultures that embrace the concept of intelligent failure more readily?

Cultural differences are huge, but I haven’t studied this area directly. Erin Meyer at INSEAD has written The Culture Map, which is a fantastic book, and identifies differences across cultures that are multidimensional and certainly attitudes toward risk and failure are one of them.

The other day someone was making the point that a manager in Japan will have a hard time saying, "Tell me the problems." Culturally, the belief is that that will make that manager look weak. And so they can't do that. But I would argue that, in that case, there is a cultural hurdle to get over before the managers in a company can do the things they need to do for the team to be excellent.

And when someone truly appreciates that some of the things that they need to do so the team can be truly excellent aren’t an easy fit with their culture, then they’re willing to change because they care enough about the goal. It doesn't mean when they go home at the end of the day, they don't know how to behave in their community or in their family. They do. But they know that this is what they do at work, because they care.

And have you seen that light bulb moment where management teams who've been resistant suddenly realised that this change will make them better?

Yes, many times. I remember a plant manager, who had a not well-educated workforce who were doing a very routine job. And he'd been a jerk. His approach has been top-down, command and control, demanding—but he hadn’t been a part of the team, helping to make the quality better.

And once he was shown that there was an alternative way of being a manager— one that would be better for results and the performance of the plant, he changed.

He’d never found it fun to be a jerk, it was just how he thought would get results. So, I think when people are really listening and they realise the logic of the approach, they will do what they can to try to change.

You note that many successful people tend to have a healthy relationship with failure, but many of us were raised to believe failure is unacceptable. How would you recommend unlearning this belief, and building a healthier mindset around failure?

Practice. Because it's very hard. I mean, I'm certainly one of those people, I had great parents, but still I'm one of those people who somehow got the message that I wasn't supposed to fail. I should get an ‘A’ on the test, I should do well at everything.

But you can't be a successful scientist or a successful academic in any field without many rejections from academic journals or in terms of student ratings or whatever. If you decide to participate in any sport or career at an elite level, or even a less-than elite level, you will experience failure.

And so, you really have two choices. One is to be perpetually miserable because you wanted something not to happen and it happened and so you wallow in that unhappiness. Or you're going to train yourself to say, okay, that's disappointing. But what do I try next?

You have to remind yourself why you care and take a pause. And then replace your spontaneous unhelpful thinking with non-spontaneous but very deliberate, helpful thinking. “So, this wasn't what I wanted, but okay, what can I take from this and what will I try next?”

You've spoken about the importance of teamwork in navigating failures. What advice do you have for student groups or project teams on fostering an environment where failure is embraced as part of the learning process?

First of all, I think student teams working on a project should always start with a kind of stage setting where they say “Okay, what kind of project is this? And has anyone done something exactly like this before?” The answer is probably, “no, not exactly”, right? So, they should remind themselves, explicitly, that this is brand new and that things will go wrong.

So, you're setting the stage for novelty, and also for interdependence. I depend on you, and you depend on me. There will be times where it will seem like you deliberately let me down, but it's much more likely that it was inadvertent, right? So, let's get on the same page about what kind of project this is. Let's get super curious about and interested in each other's ideas, strengths, weaknesses.

And, most importantly of all, let's approach this as a learning project. We obviously want to perform well, but it’s also supposed to make us better. And if you spend all your time doing the things you're already good at and not the things you're not good at, that might not be the best way to learn.

Do teams necessarily need to have a leader?

They need structure. Structure could be provided by a leader, or it could be provided by a set of rules and processes that they agree to use.

For example, they might decide to start every meeting with a check-in. Or that all meetings will be stand-up meetings and we'll do them every week and we'll share all information. If there's a decision to be made, we'll decide who owns the decision.

The easiest form of structure is to have a boss, but it might not be the best form of structure for a team project.

You talk about schools tending to have a performance, rather than growth mindset, and you also talk about Carol Dweck’s work here. What can we do at a university level to balance promoting excellence with supporting students to develop a mindset where they can learn from failure?

You know, a growth mindset is ultimately the mindset that better enables people to succeed. So we're not trading off success against a growth mindset.

And a growth mindset is not a recipe for failing the course. It is a recipe for maybe taking a more challenging course rather than a less challenging course. And maybe you'll get a ‘B+’ or an ‘A-’, instead of an ‘A+’—but that’s okay because it's a harder course.

So, you've got to wake the students up to the long-term view. Why are you here? You're here to learn and grow and develop essential skills for succeeding in life. I think it's up to Newton to encourage students to take advantage of the short time they're here to learn as much as they can so that they will be better learners—if they can learn how to learn, that will be with them their whole lives.

In addition to your work as a lecturer and an author, you work with businesses and organisations. In Czechia we face a struggle getting businesses to take higher education institutes in the social sciences, economics, management seriously, and getting  them to open their doors to us. Do you have any advice for how we can build these relationships?

Well, you know, I suppose I always had an advantage in that I was both a PhD student and a junior faculty member, assistant professor, associate professor at a school with a heck of a brand.

So when you knock on the door and say, "I'm from Harvard, and can I study you?" two good things happen. One, they think you're probably capable, which may or may not be true. And two, they think you're trustworthy. So I had an advantage.

But to generalise, I think many companies will invite you in without a guarantee that they're going to get something valuable out of it so long as they know they won't be hurt. So I think you have to figure out what you need to do to make them feel safe, so that they can open up and tell you the truth. For example, a confidentiality agreement, where you agree not to publish anything that uses their name without permission.

I also think that nowadays it's easier, given the speed of change and technology. We know we're in the knowledge era, the digital era. I think most business leaders understand that business is no longer a purely technical activity where only technical skills in finance and manufacturing matter. It's a very human thing. And the humanities, psychology, and history and literature all give people insight into the human experience that might make them better managers.

Thank you, Amy, for taking the time to talk to us.