Career & Success

Embracing Failure: How to Make Mistakes That Work

In this podcast episode, Amy C. Edmonson outlines how leaders can foster an environment where failure is viewed as a success.

April 15, 2024

| by Matt Abrahams Amy C. Edmonson

Effective and productive teams communicate safely, take risks, and aren’t afraid to fail. But fostering that environment as a leader isn’t easy.

According to Amy C. Edmondson, a professor at Harvard Business School and author of The Right Kind of Wrong: The Science of Failing Well, it starts with truly understanding failure, then promoting a culture of honest feedback, to begin embracing mistakes as successes.

In her conversation with host and Strategic Communications lecturer Matt Abrahams, Edmondson opens up about her own struggles with failure and provides tips on moving from rumination to reflection. Edmondson and Abrahams discuss the different types of failure — basic, complex, and intelligent — and their implications for innovation.

Think Fast, Talk Smart is a podcast produced by Stanford Graduate School of Business. Each episode provides concrete, easy-to-implement tools and techniques to help you hone and enhance your communication skills.

Full Transcript

Note: Transcripts are generated by machine and lightly edited by humans. They may contain errors.

Matt Abrahams: Effective and productive teams and relationships are based on the ability to communicate safely and to fail successfully. I’m Matt Abrahams and I teach Strategic Communication at Stanford Graduate School of Business. Welcome to Think Fast Talk Smart, the podcast. Today I’m excited to speak with Amy Edmondson. Amy is a professor of leadership and management at the Harvard Business School. Her research looks at teaming psychological Safety and Organizational Learning. She’s the author of the influential book, the Fearless Organization, Creating Psychological Safety in the Workplace for Learning Innovation and Growth. And she has a new book out entitled The Right Kind of Wrong, the Science of Failing. Well, welcome Amy. I look forward to our discussion today.

Amy Edmondson: I do too. Thrilled to be here.

Matt Abrahams: Excellent. Shall we get started? Let’s, alright. I first came to know you through your work on psychological safety in teams. So I’d like to start our conversation there. What is psychological safety and what specific communication strategies can we employ to foster the kind of open and supportive dialogue that you talk about?

Amy Edmondson: So psychological safety refers to a belief that you can speak up candidly with interpersonally risky content like I made a mistake, or I need help, or I disagree with your point of view. Not that those things are easy, but that you believe they are expected and even welcome. And this is a really important distinction because psychological safety has been getting a lot of attention in the last few years. And often it’s being misunderstood to mean, comfortable or not having high standards or not feeling a sense of accountability to excellence. And really it’s not that it’s simply a learning environment where you believe your voice is welcome and you don’t err on the side of holding back. So what communication helps build it? I think really first and foremost, it starts with the work. It starts with making early and frequent references to the nature of the work and why it requires us to do this.

Let’s say we’re taking care of vulnerable patients. Our processes are necessarily error prone and complex. We need to hear from you. Think about that as a message now that’s for patient care, but what if we’re developing some new technology that doesn’t yet exist in the world? No one’s ever done anything like this before. We’re going to get things wrong on the way to getting them right. It’s like all ideas, welcome. Let us know early when things aren’t working so we can pivot in a timely way. It’s about continually reminding people that the nature of the work is such, it won’t go perfectly every time. And that gives us a sense of permission to speak up, a permission to be imperfect because that’s just reality. So I think that’s the most important messaging. But right below that I would say is asking questions. It’s demonstrating a genuine curiosity like you’re doing right now. I want to ask you this question and then I’m going to listen to hear what you have to say. And of course, also to build psychological safety, you need to have productive or constructive responses. If every time you hear bad news, you bite someone’s head off. That does not encourage psychological safety. That does not encourage candor. So you have to master the pause, take a breath and say, thank you so much for letting me know. And then dig in. How did this happen? Let’s take a look together.

Matt Abrahams: What I heard there is you have to have the meta conversation about the fact that psychological safety is important. You have to be curious. You have to listen for and seek out candor and be grateful when you hear it. I have been in teams and meetings where I have felt the safety that you’re talking about. And one of the things I noticed that good leaders did in those circumstances were they acknowledged and expressed gratitude when people did act vulnerably or share mistakes. Is that a critical part of this too?

Amy Edmondson: Yes. It’s that how you respond is absolutely critical and it doesn’t mean agreeing with everything and it doesn’t mean praise for everything, but it does mean acknowledging and appreciating and especially appreciating that it’s hard. It’s hard for someone to offer a dissenting view or it’s hard for someone to ask for help or to let you know the project is not going well. That’s always going to be hard. So making that less hard by being appreciative is really important.

Matt Abrahams: Well, I appreciate that answer. What role do status, power, culture, and expertise play in all of this? Is it possible for everyone to feel similarly psychologically safe? I can imagine there might be some challenges there.

Amy Edmondson: Yes, it’s more complicated than you might think. So let me put it this way, status or power makes a difference. And when we are in an unequal status relationship or situation, there’s a boss and then there’s others who maybe report to that boss. You’ll naturally have a more difficult time speaking up candidly because of that power differential. But that does not mean we should get rid of power differentials or that it’s hopeless. What matters even more than the power differential is how it’s handled. So I did a study once with Ingrid Mhar of intensive care units, all actively engaged in quality improvement work. We had 23 of them in 23 different hospitals. They all had the exact same hierarchy. They had a medical director who’s sort of a top doc who’s very much in charge. They had nurses, they had medical residents and so forth.

And the hierarchies again, were identical, but in some, I would say even most slightly more than half of the ICUs, the degree of psychological safety between higher status and lower status people was significantly different, but in slightly fewer than half. The psychological safety was exactly the same across role groups. No difference. Now we fast forward to the end of that story. Those with the flat psychological safety pattern had ultimately an 18% improvement compared to their counterparts in morbidity and mortality. So same hierarchy, but what was different? Well, what was different was how those with the power, in this case, sort of the top doc, the medical director, how they handled their status. And these were the ones who actually did things I didn’t tell them to. They were just thoughtful enough to do it. Very similar to what you were just asking me about. So they would routinely say things like, I might miss something, I need to hear from you. Or they’d just reach right out and say, Matt, what did you see last night? I think you were on with patient, so-and-So or they would say things like, thank you so much for bringing that to my attention. They were just in a sense, thoughtful, wise clinicians who did those things and that made the natural differential just go away. So I think it’s really important to say that differences in status, power, culture, expertise matter, but what really matters is how we handle them.

Matt Abrahams: Right? And literally in situations of life and death, as you just described in those positions of power and status, by acknowledging, rewarding and being open to those who have less, it’s helpful. Do you have any research or advice on people who have lower status, who might not feel comfortable sharing their positions for various reasons, what they can do to help contribute to the overall psychological safety of a group?

Amy Edmondson: So I don’t have actual research in the same way that we studied what the leaders were doing and why it mattered. But what I can say is I think there’s a fair amount of anecdotal evidence to suggest that any team member who asks a thoughtful question and then listens carefully to the response, whether that’s of a superior, a peer or a subordinate, is in that moment giving others a little window of an opportunity, a stage, a little opportunity to express themselves. And that demonstrates respect. It’s contagious, and it’s not that hard to do if we take the same messaging that you asked me about earlier and say, wait a minute, do you have to be the boss to do it? I think the answer is no. Anyone can ask a good question. Anyone can be appreciative of what someone else says. Anyone can call attention to the fact we’ve never done anything like this before. What ideas do people have? And they will be appreciated, seen as helpful, and nothing bad happened to them, so others will follow suit.

Matt Abrahams: Excellent. So we all can contribute.

Amy Edmondson: And I think that’s really important because psychological safety is not sort of a switch that you turn on and now we have it. It’s more something that is enacted day by day, day by day. It’s enacted by each of us taking small interpersonal risks to be the risks of being candid. And it’s enacted by nobody getting really mad at us or kicking us out of the team as a result of our doing that. So we’re always co-creating it. I think the higher you go, the more of an outsized influence you have on that climate, but we all influence it.

Matt Abrahams: It’s something that has to be worked on continuously. Part of feeling psychologically safe is feeling okay, making mistakes and admitting them. This idea of failure is the focus of your new book, right? Kind of wrong. Can you talk to us about unproductive failure versus productive failure? And can you share your three archetypes of failure for us?

Amy Edmondson: First, let me say that psychological safety is absolutely about feeling okay, admitting mistakes. It doesn’t necessarily make you feel happier or okay, making mistakes. I mean, I think all of us would rather not make mistakes. We’re all fallible human beings, we’ll all make mistakes, but it’s more about speaking up about them. So I’ll say that now. I think it’s important to distinguish between mistakes which are deviations from known practices that will achieve the results you want, whether a recipe, a playbook, what have you, and failures, some of which are caused by mistakes, but some of which are the absolutely unpredictable negative outcomes in new territory. I think people use the two terms interchangeably. Failure is a bigger term, mistakes is a more precise term. And to the three archetypes of failure, let’s start with basic failures, which are failures caused by a mistake. I put salt in the recipe instead of sugar by mistake, and it’s lousy fine.

And by the way, mostly avoidable complex failures are failures that are caused by multiple factors coming together in just the wrong way. The perfect storm and main difference with basic failures is that any one of the factors on its own would not have led to a failure. It’s the fact of them all coming and lining up almost with an element of bad luck there that creates a bad outcome. And then neither of those are good news. They’re both things we can learn from, and they’re both things we can work hard to try to reduce in our organizations. The third kind of failure are intelligent failures, and those are the undesired results of thoughtful forays into new territory. So intelligent failures have in common that they’re in pursuit of a goal, they’re in new territory. There’s no playbook yet that you could just look up on the internet and follow. You’ve got a hypothesis, you’ve got good reason to believe this might work. And finally, the failure is no bigger than necessary to get the knowledge that we need to make progress. So you readily recognize that as sort of the activity of scientists or inventors or other people as a living, doing things in new territory, creating new things.

Amy Edmondson: Entrepreneurs and technologists in general are routinely going into new territory. So they have to live with the reality that you cannot get it right every time because there isn’t that playbook. And so you have to be willing to experiment, but you want to experiment thoughtfully. You don’t want it to be just random stabs in the dark. You want to have, as I said, good reason to believe it might work. That’s kind of the Silicon Valley ethos.

Matt Abrahams: Thank you for delineating the different types of failure. How can leaders communicate about failures and mistakes in a way that encourages learning and resilience rather than fear and blame?

Amy Edmondson: One of the chapters in the new book is about the importance of context, the importance of situational awareness, and I think that’s where leaders can play a really important role is by calling attention to context. Now, context can mean a great deal of things. I’m primarily interested in two dimensions of context. One is how much uncertainty is there? Is this a situation where there’s a nice recipe, let’s follow it and it will work every time? Or is this a situation where we’re truly on the bleeding edge of some field and we have no choice but to experiment? So that’s low to high uncertainty. The other dimension is what are the stakes? What’s at stake here? Whether from a human safety point of view, as in healthcare or financially in many businesses or reputationally. And if the stakes are very high and yet it’s high certainty, please follow the protocols vigilantly carefully.

If the stakes are relatively low, you’re behind closed doors and you’re just experimenting in a lab, the uncertainty is high. Go to town, take bold experiments and fail fast, learn as fast as you can. And of course, if it’s high stakes and high uncertainty, you’ve got to find a way to break that down, to get to little pieces of it so that we can learn more. We certainly don’t want to take undue risks to safety, finances, or reputation. So the job of the leader is to say, what are the stakes here? And then help people think about it and act accordingly. I have seen both errors. I have seen people behave carelessly in very high stakes, uncertainty situations leading to such things as NASA shuttle disasters and worse, and I have seen people behave overly cautiously in domains that caused for playfulness where the stakes were remarkably low, where behind closed doors were experimenting, and yet there’s that human desire to want to get it right the first time, leading to unhealthy hesitation and unwillingness to really experiment.

Matt Abrahams: What I’m taking away from several of your answers is that all failures are not equal, and we need to really be thinking about the different types of failure and leaning into some and then making some very conscious choices about others. I want to share an experience I had when I was an operator working in high tech. I worked for a company that had failure Fridays, and we would all convene for lunch and you could nominate yourself to share a failure. You had that week and the senior leaders would vote on the best failure. You actually got an award back in the day. It was enough to take a partner out to a meal and if you had kids to pay for babysitting. And the idea there was that if we weren’t pushing the envelope, we weren’t growing. And it was a teaching moment because the rule was you could never have the same failure twice win. So we all had to learn from each other. And I remember this vividly, and this was decades ago, this to me seemed like a really powerful way to encourage the kind of things you’re talking about in terms of failing forward.

Amy Edmondson: Exactly. And when you had those kinds of rituals, when organizations or teams that have those kinds of rituals either intuitively or explicitly know, the failure Friday is not about coming together and saying, I texted and drove this week and I got into a car accident. No, that’s a basic failure. Human error led to an undesired outcome that was preventable. And maybe there’s some value in the reminder to others not to do that, but that’s not what they were talking about, right? They were talking about you tried something bold and on purpose, meaning in pursuit of the company’s value proposition, and a lasset didn’t work or the client didn’t buy it. But it was a bold and lovely idea, and thank you for sharing it because now A, I won’t do the same thing. And B, you’re sending that wonderful message that this is how we work in uncertain territory.

Matt Abrahams: Yes. And the motto of that company was scary, fun. Which fits right into what you’re talking about. I want to look at a more serious issue. You discuss the notion of the unequal opportunity to fail that exists in many groups and organizations. Can you describe this a bit and share what can be done? Not everybody has the same right to fail in some ways.

Amy Edmondson: So much of our conversation thus far has been about the modal person, let’s say in Silicon Valley or elsewhere, who gets this stuff that here’s this, yep, we better innovate, we better try things. Not everything will work and your team will react well and all will be well. But what if you are a way underrepresented minority with respect to the role that you occupy? What if you’re a senior executive and no one else looks like you or you are a consultant in a client’s office? Again here you do not feel you have that same permission to fail intelligently because there is an added weight that is thinking either consciously or not. When I fail, it reflects badly on my group, on my identity group. Other people will not have the same opportunities I have if I don’t get this right. And that’s incredibly unhelpful for them and for the world because it’s going to make them more risk averse, understandably, because they understand how the human process of stereotyping works. So what can we do about it? Right? I think that most important, maybe the only thing we can really do about is call attention to this as a bind, call attention to this as a problem and commit out loud together to making it less so to making sure that people from underrepresented groups are having an equal opportunity to shine, and that we are just as intent on celebrating their intelligent failures as anyone else’s. And of course, try to do what we can to make them less rare in the positions.

Matt Abrahams: Absolutely. I think that’s wonderful advice. And I think also trying to minimize consequences if and when those failures happen is critical. So Amy, before we end, I’d like to ask you three questions. One I make up just for you, and two, we ask all of our guests, are you up for that?

Amy Edmondson: Yes.

Matt Abrahams: I’m curious, when it comes to failure in your own life, what have you learned and how have you best leaned into the failure that you’ve had?

Amy Edmondson: Well, I’ve learned that I do not spontaneously have a healthy response to my own failures, and I don’t have a response to either intelligent failures, which as a researcher I should because I get that stuff intellectually, but still, emotionally, I would’ve rather been right than wrong. So I will go into a kind of emotional tailspin before I catch myself and correct that when say the hypothesis is not supported by data, and I’m even worse at responding to the basic failures, the ones that were absolutely unequivocally caused by my mistakes, I can easily end up ruminating rather than reflecting. So I have to catch and correct myself all the time.

Matt Abrahams: Thank you for sharing that. I mean, you are human and these things can be frustrating and an emotion plays an important role. Let me ask question number two. Who is a communicator that you admire and why?

Amy Edmondson: There’s so many good answers to that, but the one that I thought of is Nikolai Tangen, and he’s the head of the Norwegian Growth Fund financial fund that supports many activities in Norway. But why I find him such a compelling communicator? He’s got a podcast where he interviews leaders, occasionally academics, but mostly CEOs of companies that he is investing in. He’s a great question asker. He asks good questions, but what he’s also doing is allowing them to communicate their wisdom and their messages. So he’s putting them in the spotlight in enormously thoughtful and productive ways to get, ultimately, I think his messages across using this vehicle of the podcast, much like what you’re doing.

Matt Abrahams: Well, I love that you picked a podcaster as somebody you admire, and the ability to tee people up to share their experience and knowledge is a true art form. So our final question, what are the first three ingredients that go into a successful communication recipe?

Amy Edmondson: I’m going to say clarity, caring, and commitment. So clarity is take the time to think through what is it I’m trying to convey? Be clear about that yourself so others can be caring is remind yourself you really do care about them and the work and what it is we’re trying to do. And commitment, I think is the passion that you bring to it that conveys that you are all in

Matt Abrahams: Clarity, caring, commitment, A wonderful recipe. Thank you so much, Amy. I really appreciate the many specific actionable takeaways you provided around how to develop and support psychological safety and how we can all embrace productive failure. And I wish you best of luck with your new book, right? Kind of Wrong, the Science of Failing. Well,

Amy Edmondson: Thank you so much for having me, Matt. It’s been really fun to talk to you.

Matt Abrahams: Thank you for joining us for another episode of Think Fast Talk Smart, the podcast from Stanford GSB. To learn more about psychological safety and failing successfully, please listen to episode 26 with Sarah Soule and episode 112 with me. This episode was produced by Jenny Luna, Ryan Campos and me, Matt Abrahams. Our music is from Floyd Wonder. Please find us on YouTube and wherever you get your podcasts. Be sure to subscribe and rate us. Also follow us on LinkedIn and Instagram and check out for deep dive videos, English language learning content, and our newsletter.

For media inquiries, visit the Newsroom.

Explore More