972: Amy Edmondson on How to Fail Well

Pete Mockaitis
Amy, welcome back.

Amy Edmondson
Great to be back.

Pete Mockaitis
Well, I am super excited to hear your wisdom and talk about failing well. And I’d like it if you could kick us off with one of your favorite failures, personally or professionally?

Amy Edmondson
My favorite failure has to be the time, as a second year PhD student, I had a hypothesis that better teams would have lower medication error rates. I finally got the data after six months of data collection, put it all together, ran the numbers, and, alas, I had failed. My hypothesis was not only not supported, it was 180 degrees wrong.

In other words, the data suggested that better teams measured by a validated team survey instrument had higher error rates, not lower. So, that was just, you know, my dreams of publishing a paper evaporated in a moment as I looked at the screen, and I felt quite despondent about it. So, that was a failure, right? There was no question about it. But of course, it’s a favorite because, ultimately, it pointed me in the direction of a success.

Pete Mockaitis
Well, tell us about how that unfolded.

Amy Edmondson
Well as I tried to figure out what this might mean, “What does it mean when a valid team survey assessment data suggests that these are really better teams, they’re better led, they’re more engaged, they have higher quality relationships? What does it mean when those kinds of teams appear to have higher error rates than their counterparts?”

And it suddenly occurred to me, and what I later thought of as a blinding flash of the obvious, that maybe the better teams aren’t making more mistakes. Maybe they’re more willing to report them, maybe they’re less likely to cover them up, to keep them secret, to shove them under the rug. And the more I thought about that, the more possible, even likely, that became. And, of course, it’s hard to talk about mistakes at work. It’s hard to admit you’ve done something wrong. It’s hard to ask for help when you don’t know quite what to do.

And because of that, what I increasingly started talking about, interpersonally risky nature of those behaviors, it would, in fact, be, at least, it could explain the unexpected findings. It could be an alternative explanation, that the dependent variable, the error rates, was actually a faulty measure. It was a measure that was subject to reporting bias. And just having that insight, of course, wasn’t enough to prove it or to do anything else with it, but it led me down a path of trying to understand whether teams, in fact, have different interpersonal climates, especially around something fraught like error.

And if so, would that affect their learning behaviors? And if so, would that affect their performance? And so, that led me down a road of doing a very different kind of research, which was to sort of explore those possibilities. Later, I called that interpersonal climate, with the input of a peer reviewer on a paper, psychological safety. And so, that was sort of the birthplace of a, then, thriving research program on team psychological safety, which turns out to be a very powerful predictor of team performance in a variety of industry contexts.

Pete Mockaitis
That’s right. Psychological safety is huge, and you are the queen of it, and thank you for sharing those insights with us in our last conversation. So, yeah, now let’s dig into some of the goodies you shared in your latest book, Right Kind of Wrong: The Science of Failing Well. Any particularly striking, surprising, counterintuitive discoveries that you made along the way and put forward in this book?

Amy Edmondson
Yes, you know, the failure I just described was what I do call in the book an intelligent failure, and it’s intelligent because, and in research there’s many intelligent failures if you’re doing it right, which is that you are pursuing a goal of developing new knowledge. It’s in new territory. We don’t yet know what will happen. You’ve got a hypothesis, like you’ve done your homework to find out what we know, what we don’t know, and you have a good idea about what might work, and the failure is no bigger than it has to be.

You don’t spend your whole research budget on one experiment. You sort of do it thoughtfully so that you can learn in a reasonably efficient way. And so, science is full of intelligent failures, of course. But I think the surprising thing that I learned through all of this research is that, as human beings, we respond similarly to intelligent failures as to what I’ll call basic and complex failures, the preventable kind, meaning our emotional reactions are similarly negative and unhelpful.

As I started telling my story of my research failure as a second-year graduate student, the emotions I felt were downright catastrophic. I was starting to envision I’ll have to drop out of graduate school, I’m no good at this, I failed, you know, crazy stuff, which is just factually wrong-headed, but that’s where your brain goes. A failure is always disappointing, and when you have one, you can easily, or at least I can, spiral into unhelpful, unhealthy thinking.

And so, I think an opening surprise or insight is that, even though some failures logically are wonderfully helpful in making progress in new territory, we are vulnerable to having an emotional reaction to them that, then, precludes us from learning what we need to learn from them.

Pete Mockaitis
Yeah, that is huge, and I want to dig into that in some real detail. But first, could you maybe make the grand case associated with failure being okay, or the main thesis, or the takeaway from Right Kind of Wrong?

Amy Edmondson
Yeah, so it’s not actually as simple as “Failure is okay.” It is as simple as intelligent failure is not only okay, it’s praiseworthy. If you are willing to take risks, and we all must be in order to make progress in our lives and in our jobs, if you’re willing to take risks in new territory, you will face some failures along the way, otherwise, they weren’t risks. If everything you try just goes perfectly, you’re probably not stretching very much, and that’s not a way to be awesome at your job.

So, those are the kinds, you know, the intelligent failures in new territory are indeed the kinds we’ve got to sort of develop the muscles to welcome them, and be not only not despondent about them but actually genuinely delighted by the opportunity to learn and sometimes pivot. But it is equally a book about pulling together what we know about the best practices for preventing preventable failures. One kind of failure I identify is a basic failure, which has a single cause, usually human error, or deviation from the process or protocol or recipe, and those are theoretically and practically preventable.

And so, for instance, in a new job, if you are looking around and you’re not quite sure what to do, and you don’t feel comfortable asking someone, and then you do it wrong and it leads to a failure, that’s a basic failure, and those we aren’t so excited about. It’s a book about failure but it’s really a book about success because the idea is, “Let’s do everything we can to execute well in known territory and to explore well in new territory.”

Pete Mockaitis
Well said. Well said. Well, let’s continue this typology or categorization here. We’ve got basic failure, a deviation from the recipe. My wife and I, we joke about this all the time. She’s like, whenever I cook something, and she said, “Wow, that was great. What’d you do?” And I said, “I did exactly what the recipe told me to do, because I have learned from many, many basic failures, that whenever I think I know better, or this is no big deal, how about I do this little twist, it almost always ends poorly,” unless sure enough, I’ve done the recipe many times, and I thought “Okay, I know how this works. I’m going to put a little extra of this in there because I know that I like this.” And it usually works out okay in those circumstances, so I like the word recipe in there.

Amy Edmondson
ecome an expert in it, and then you’re in a good position to experiment.

Pete Mockaitis
All right. So, we’ve got our basic failures. And then what else?

Amy Edmondson
Complex failures. So, complex failures are quite simply the perfect storms that bother us and show up in our lives and our organizations. They are the failures that are caused by a handful of factors coming together in just the wrong way, any one of which on its own would not lead to a failure, but the fact of their coming together. Let’s say you accidentally set your alarm clock for p.m. rather than a.m. I’ve done that.

So, maybe you wake up a little late but you still can get to that important meeting on time, but not if you also hadn’t realized your gas tank was hovering on empty and there was a tractor-trailer accident on the highway, five little things happen, none of which on their own would have led you to be late for this important meeting, but they all came together to lead to this complex failure.

Our lives are complex, our world is complex, so complex failures are kind of on the rise, which is depressing. But I think the silver lining of complex failures is that all you really have to do is catch and correct one of the things. Now, some of them are external. There was nothing you could have done about the tractor-trailer, except if it was a really important meeting, you leave a lot of buffers in there. So, then you would have taken extra care to have set the alarm accurately.

But the beauty of them, I mean, the bad news about them is that they’re sort of everywhere and they can just happen when we’re not really at our very best. The good news is, when we’re vigilant, we can catch and correct, and if you sort of catch and correct any one of the factors, you usually can dodge the failure.

Pete Mockaitis
Okay. And talking about the emotional piece, those are the situations, when you can feel it progressively melting down, that just drive you bonkers.

Amy Edmondson
Yeah, likewise, you sort of see it in slow motion heading your way, and maybe it’s too late to change it, but maybe there’s still time, or at least you can put in a plan B. You can make that phone call and explain and maybe reschedule.

Pete Mockaitis
All right. And we have additional categories?

Amy Edmondson
Nope, just the three: intelligent, basic, and complex. And there’s a bright red line drawn between intelligent and the other two because the intelligent ones are the ones that we need to have more of, we need to welcome, we need to just force ourselves to like them. And the other two, we need to kind of sit up straight, pay attention, and try to prevent as many as possible.

Pete Mockaitis
All right. Well, I was going to ask, sit up straight, pay attention, that doesn’t sound hard.

Amy Edmondson
No. That doesn’t sound…yeah, that’s not complete, is it?

Pete Mockaitis
Yeah. Are there some basic protocols, or checklists, or things you recommend to help us prevent those?

Amy Edmondson
Yeah, absolutely. And as I start talking about these things, the worry I always feel is they sound so basic, they sound so simple, and they are. And, unfortunately, they need to be used with intent. So, a checklist, for example, it’s a brilliant tool. It helps us remember to do some of the right things, when maybe it’s just a little complicated, and it would be easy to miss something, turning the anti-icing on, for example, taking off on a wintry storm day, and so we have these structures and tools to help us do the right thing.

But if you have those tools and structures in your life, or in your organization, but you’re using them, roughly speaking, in your sleep, just not really paying attention, not using them with intent, then they don’t help at all, right? If you’re sort of going through the motions of using a checklist but you’re really not paying attention, and you have just a habit of check, check, check, check, check, without your brain in the game, they won’t work. So, no tool is good enough on its own to overcome kind of the vagaries of human nature and lack of discipline.

Pete Mockaitis
And when you talk about good enough, I’m thinking about some people think that they are too good for the checklist.

Amy Edmondson
But that’s another point, isn’t it?

Pete Mockaitis
A Checklist Manifesto, or something, talk about surgeons or whatever, who were reluctant to lower themselves to simply be in checklists.

Amy Edmondson
Right, and that’s because they had the wrong mental model. So, even the wonderful Atul Gawande, who wrote the book, The Checklist Manifesto, I remember hearing him interviewed on NPR, and it was so brilliant because he was asked by the interviewer, “You’re a Harvard surgeon, and so on, a beautiful writer, do you use a checklist yourself in the operating room?”

And he said, “Well, in writing this book, I knew I’d be asked that question, so, of course I had to force myself to use them, even though,” he claimed honestly, “I didn’t think I needed it, right? That’s sort of for mere mortals, but not for me.” I mean, he didn’t say it that way, but it’s exactly, as your question, put it. And so, he said, “And I discovered something.”

He says, “Never does a week go by that that checklist, that I forced myself to use because people like you would be asking me this question, never does a week go by where that thing doesn’t save my process in some way. So, in other words, it turns out I am mortal. I am vulnerable to making mistakes and this very simple, simple tool, this checklist, has saved me on numerous occasions.”

Pete Mockaitis
And I’m curious, as we think about professionals and basic failures, what are some basic failures you see over and over and over again that might be well-suited to a checklist or some sort of a tool along those lines?

Amy Edmondson
If we all think about our experiences as customers, or as patients, just on ordinary annual checkups or something, trying to get the schedule, or trying to, you know, there’s so many organizations that don’t work as well as they should, that you can look around and say, “This could be better with just a tiny bit of scaffolding to allow the recurring, repeating activities to be a bit more programmatic, a bit more supported.” And I know that’s very abstract, right?

Pete Mockaitis
But no, I think that’s a huge tool right there in terms of, you know, it’s funny when we instituted this, when I’m making courses, I have someone, his name is Ian – shoutout to Ian – and we’ve given him the title of Neutral Ambassador of Learning. And the idea is that Ian had nothing to do with producing any of the materials anywhere around the chain, but his role is to step into what we’ve created, and be that first observer, that first voice, and tell us, “All right Ian, what’s wrong? What chapter title was off, wrong? Which bits of the content seemed incomplete or confusing?”

And it’s fantastic! He surfaces all sorts of things, like, “Oh, we were so close that we didn’t even notice.” But with every customer experience kind of a situation, you’d think it’s like, “You know, if you actually tried to fill out this form yourself once, that you’re making us fill out thousands of times, you’d realize that I need more than an inch and a half to write an address. You’d know it on your form.”

Or, that, “Maybe there might be a means by which you can take my name and birth date from one medical form and auto-populate that into the nine other medical forms I have to go through, to have this doctor’s appointment.”

Amy Edmondson
Because if you can do that, you are reducing the likelihood of error, of just simple human error… You probably won’t screw up your birth date. Although sometimes, on my birthday, I screw up my birthday every year because I just automatically write, once I’ve started with my birthday, the year follows naturally.

And so, there I am writing a year from many, many decades ago in the form, when it’s like, “Oh.” It’s just a simple brain malfunction. Right, you write the actual birth date, but yes. So, now I’m thinking, “What are the failures that could’ve been avoided with a simple checklist or making the thing a bit more programmatic?” For instance, in the world of podcasts, I have been on at least one where there was a failure to press record. It’s not catastrophic, nobody dies, but, boy, it can feel almost catastrophic.

Pete Mockaitis
That happened to me once. I felt so embarrassed.

Amy Edmondson
So bad, where you’ve spent all this wonderful time and energy, and then, “Oops. Oops.” It was great for us but the rest of the world would not hear this one, and doing it a second time is just not…but that is one of those errors, by the way, that you learn from and rarely do it a second time. But it is certainly possible to never do it the first time also.

Pete Mockaitis
And I love it, in the software now, we’re on Riverside, and I believe it does this too, it’ll now tell you. If we’re chit-chatting for too long without record, a notification will come up, “Notice, you’re not actually recording now.” So, that’s cool though.

Amy Edmondson
Which is good! Which is good, right? It’s sort of a little check. It’s a check that’s built into the system.

Pete Mockaitis
I dig it. Okay. So, then let’s talk about these intelligent failures. You say there are a few criteria that we need to check in order to say that something is an intelligent failure before we could just give ourselves credit, and say, “Oh, that was an intelligent failure. It’s okay.”

Amy Edmondson
Yes. So, first and foremost, it’s in pursuit of a goal. I’m not against just messing about, playing with resources, but that’s another category of activity. That’s play. But if something is an intelligent failure, it is the case that you are pursuing a goal, whether that is a better recipe, or a life partner, or an innovation in your job at work, but it’s something that you’re hoping to accomplish and improve, and it’s arguably in new territory.

It’s not possible to just get the answer for what will happen off the internet. There’s no way to get the new knowledge you are trying to get to make this progress without trying it, without the experiment, and you have a hypothesis where you’ve done enough thinking and homework to have a good sense that this might work, “I know it’s not a slam dunk, but it might work.”

And then, finally, fourth criterion is the failure, if it happens, is no bigger than necessary. It’s just as big as it has to be to give us the new knowledge that we need. And I always say there’s sort of a bonus fifth criterion, which is you take the time to learn from it. And so, those four criteria – pursuit of a goal, new territory, done your homework, no bigger than necessary – can apply to all sorts of realms. They can apply to a blind date. They can apply to improving that cookie recipe. They can apply to an assignment in your job.

Pete Mockaitis
That’s great. Well, now, Amy, I’d love it if you could give us some fun examples of taking this mindset and approach, and doing it with gusto. What came to mind for me was I’m thinking about Tony Hsieh’s Zappos story, is that one of the first things he had to test was, “Are folks even willing to buy shoes online?” This was back in the day. He didn’t know yet. So, he partnered with a shoe store and said, “Hey, here’s the deal. I’d put a photograph of the shoes in the store, and every time I sell one online, I’ll come here and buy it from you, and then ship it to them.” And so, that’s wildly inefficient from, like, a profit perspective.

Amy Edmondson
Right, but it’s brilliant. It’s brilliant.

Pete Mockaitis
But from learning, yes. It nails it.

Amy Edmondson
Because imagine, I mean, compare that little experiment, expensive experiment, but compare that to setting up a whole warehouse and filling it with shoes of all sizes, and then sort of waiting and hoping. I mean, since we don’t know if customers will buy them online in the first place, what if they don’t? Well, we have just spent tens of thousands of dollars filling up our little warehouse here with shoes, and it failed. That’s not so good. Whereas with him, if it had failed, then all he would’ve lost was a little bit of his time, and maybe he would’ve disappointed this nice shoe store, but no harm done, really, right? So, it’s a beautiful example.

Pete Mockaitis
Absolutely. Well, so please, give us some more. As we’re thinking about professionals who want to have more intelligent failures more often, any cool tales that illustrate how this could be done well in practice?

Amy Edmondson
One of the intelligent failures, and it’s not professional, it’s personal, that I love, intelligent failure stories, is my high school friend, Laura, who in her 40s, decided she wanted to learn how to play ice hockey, which is kind of crazy if you think about it, but, anyway.

Well, she lived in New Hampshire, so she thought it was really one of the things that people around her did all the time, and so it seemed like a good social activity and so on. And I think, it almost goes without saying, that her first forays onto the ice were failures, and it was intelligent because, again, she had a goal to join this league, to have some fun. She had done as much as you can do. She’d skated as a child, but in figure skates, not hockey skates. She knew enough to know that she sort of liked sliding around on the ice, and she didn’t try to join the Bruins or anything. She joins this sort of local women’s league.

And certainly, those first few weeks would not have made her a very valuable team member but she got better and really fell in love with the sport, and it remained an important part of her life now. So, that’s where someone is willing to take the risk of doing something new that might not go well. It’s not new territory for the world, but it was certainly new territory for her in her life.

I suppose in business, you bring up Tony and Zappos, but earlier than that, Amazon, obviously buying books online is not as challenging as selling shoes online, but really, “Would enough people go? Could you make it? Could you make a company work out of that?” And, of course, we later understand that Jeff Bezos had much bigger dreams in mind, but he didn’t start with, “Okay, I’m going to become the retailer of everything,” and assume that’s going to work.

He started with, “Well, let’s see if we can sort of create a little bookstore online, and drum up some enthusiastic customers, and then let’s see if we can extend some of those customers into buying other things. And then let’s see if other customers, who maybe don’t buy books that often, will come and buy other things because our operations are now so good.” So, that would be another story of one risk at a time.

Pete Mockaitis
Beautiful. And then let’s talk about this emotional piece. When we hit a failure, we can think, “I’m no good at this,” and we’d have an emotional reaction that prevents us from doing the learning, doing the persisting, doing the optimal response that keeps us moving forward. Boy, in the heat of battle, Amy, how do we address this stuff?

Amy Edmondson
So, this is why I spend some time in the book talking about becoming more self-aware and really more self-reflective and able to pause our unhelpful, unhealthy thinking and redirect it. So, I described my unhealthy, unhelpful thinking in response to my research failure years ago, and I had to learn, and we all have to learn, to kind of talk ourselves away from the wildly emotional and really inaccurate response to the disappointment of failure and rethink it, and sort of reframe it.

Reframe it from “Oh, this is awful. I’m going to have to drop out of my PhD program,” to, “Oh, this is disappointing. I wonder what it might mean?” Now, that’s a shift, or it’s a major cognitive shift, but it’s not unheard of. It doesn’t seem impossible to our listeners, I imagine, that you can practice and then learn to catch yourself, and correct your exaggerated thinking, and turn it into more scientific thinking that brings cooler, more logical thoughts, that also cool the hot emotions.

Pete Mockaitis
Okay. Could you give us an example of that in practice?

Amy Edmondson
Well, I guess an example in practice is I’m going to go back to this really important meeting, maybe it’s a job interview, and I experienced that little complex failure and I’m really unprofessionally late. I show up late. In fact, I show up so late that they’ve moved on to the next candidate. And my sort of human being’s response might be, “This is awful. I’m not going to get the job. This is the end of my life. No one’s going to hire me, and I’m going to starve to death.” Well, no, right?

So, that’s what the psychiatrist, who I talk about in the book, Maxie Maultsby, would call awfulizing or catastrophizing something that is, indeed, not great, not a great performance, but not the end of the world. And so, once you recognize that, that little emotion taking hold, that kind of your amygdala gets hooked and, says, “This is just terrible,” you pause, and you say, “No.” You force yourself to say, “This is inconvenient.”

That’s my favorite word, right? It’s inconvenient. Because, yep, it’s inconvenient, and it’s not the end of the world. And either it’s, “I’ll make amends, I’ll make an apology, maybe get a second chance,” or, “I will definitely not do this again for my next interview.” You figure out. It’s a shift from a backwards-facing, highly emotional, “This is terrible,” to a forward-facing, “Okay, that wasn’t ideal. What will I do differently next time?”

And this is how we continuously cool our emotions down, but we also force ourselves to keep learning, keep getting more thoughtful, keep getting more disciplined and wiser, and, ultimately, more creative as well, because we’re more willing to take risks because we know that the downside of the risks won’t be catastrophic.

Pete Mockaitis
That’s good. That’s good. And if we’re having trouble even beginning to think clearly and rationally and calmly about the matter, any other pro tips? Is it sleep? Is it walking? Is it cold water? Is it any of that?

Amy Edmondson
Oh, yeah, all of the above, depending on what you like. So, I would say you start with the proverbial and the real deep breath. That deep breath will interrupt the automaticity of the thinking. But, for me, certainly one of the things that I habitually have relied upon is go for a run. If I’m really stuck on something, and I just change the scenery, literally by put some running clothes on and go for a run, it completely changes how I’m thinking. It seems to put things more in perspective, and so exercise, that’s one, but anything that interrupts.

You could have a checklist of questions to ask yourself. It could be along the lines of, “Okay, what truly are the consequences of this? And what did I learn? And why is this now something that I’m able to put to use going forward?”

Pete Mockaitis
Lovely. Well, Amy, tell me, anything else you want to make sure to mention before we shift gears and hear about some of your favorite things?

Amy Edmondson
So we’ve just been talking about sort of self-awareness, self-reflection, self-discipline. Another key domain is context awareness. Be more mindful and thoughtful when you’re in a dangerous context. I know that’s obvious, but it’s truly violated all the time. People don’t put their safety equipment on when they should have, or they text and drive.

So, develop a habit of thinking quickly and clearly about the context, “What are the stakes here? What are the risks? Financial? Reputational? Safety? And what’s the uncertainty?” And act accordingly. Have lots of fun experimenting when the stakes are low. Be very, very thoughtful and vigilant when the stakes are high.

Pete Mockaitis
Alrighty. Now, could you share a favorite quote, something you find inspiring?

Amy Edmondson
One of my very favorite quotes is Viktor Frankl, which is, “Between stimulus and response, there is a space. And in that space lies our opportunity to choose. And in that choice lies our freedom.”

Pete Mockaitis
All right. And a favorite study, or experiment, a bit of research?

Amy Edmondson
I think my favorite bit of research is in the research on the fundamental attribution error by Lee Ross at Stanford, where they show in some sort of laboratory experiments that we are relentlessly willing to attribute others’ shortcomings to their own personality defects or character defects. But our own, we will very quickly and naturally see the situational forces at work. So, once we know that we’re likely to do that, once we truly internalize the idea that we will spontaneously do that, then we can step onto the road of becoming a better person, and giving others the same benefit of the doubt we often give ourselves.

Pete Mockaitis
All right. And a favorite book?

Amy Edmondson
I have many long-standing favorite books, but I’m in the middle of one that right now is becoming a favorite book, which is The Road to Character by David Brooks.

Pete Mockaitis
And is there a key nugget you share that really seems to connect and resonate with folks; they retweet it, they Kindle book highlight it, an Amy Edmondson bit of wisdom?

Amy Edmondson
Choose learning over knowing, and it’s an active choice. We have to choose it because our habitual cognitive response is to feel like we know, feel like we see reality. We have to get curious. We have to choose learning over knowing.

Pete Mockaitis
All right. And if folks want to learn more or get in touch, where would you point them?

Amy Edmondson
I guess AmyCEdmondson.com.

Pete Mockaitis
Okay. And do you have a final challenge or call to action for folks looking to be awesome at their jobs?

Amy Edmondson
Yes, take more risks.

Pete Mockaitis
Okay. Amy, thank you. This has been lots of fun once again. And I wish you many delightful failures.

Amy Edmondson
Thank you. And you, too.