Transcription
So, it's wonderful to be able to be here and have this conversation. What Daniela did not say in her introduction is actually this is a very special moment. So, on one hand, there's the joy of having your student introduce you, and Daniela's in my class, as she said. And then on the second hand, it never happened before, and you may not know, Amy Edmonson is actually my dissertation doctoral advisor. So, if you had told me years ago that I would ever be on sort of a stage with her doing something like that, I wouldn't have believed it. So this is actually quite a special moment and to be able to have the conversation that we're going to have today.
So, what Amy and I decided is that we would sort of pretend that we weren't in a big auditorium and we would just kind of have a conversation. So, you want to pretend? Yeah. Okay. So let's pretend. Um, that said, I'm going to ask her to do a little bit of background. So, I've obviously known her a very long time at this point. We were both very young um at the time we started together. Amy, why are you studying failure? Like, I mean, I feel like you're an optimistic person. I studied with you psychological safety and other things, which I think has a more positive bend, but now it feels like you've taken this turn in your career and all of a sudden you're all about failure. So, why are you focusing on failure now?
Gosh. Um, that's one way to look at it. But let, let me, let me say that I, I, I don't know whether to say I've always been focused on failure, which is true. Um, or to put it a different way, which is also true, which is I came to graduate school to study organizational learning. And so I was, I was um, really enthralled with the idea. I'd read Peter Senge's book, The Fifth Discipline, and I thought, we've got to create learning organizations. And pretty early on in my PhD program um, I made the dramatically insightful connection between mistakes and learning um, and failures and learning. And and not just being funny about that. But but truthfully, I had an opportunity to study adverse drug events in hospitals as a, and I realized, you know, why would I say yes to that? I would say yes to that because it seemed to me that if organizations are really to be learning organizations, one of the things they must learn from are the failures and mistakes. We're told that in our lives, right? So, so I wanted to understand how they do it. What are better and worse practices for learning from the failures and mistakes that will invariably happen at some point. So, it is, I guess, has an optimistic bent in that there is this premise that we can help organizations be better at learning, adapting, innovating, changing as needed to thrive in a changing world.
Okay. So, it's more positive than failure. Yeah. But now, as you've moved forward in thinking about failure, you've pushed us in the direction of starting to delineate amongst them. Right? So, you, you started by saying adverse drug events, mistakes, right? In common nomenclature, we just talk about mistakes or errors, but I think you're pushing us in this direction now to think about, well, there's actually variation in failures, and that might actually have real implications. So, maybe for me and others, maybe you could just, like, what are these? Do and do I need to pay attention to these? Like, why are these failures?
Okay. Right. So, all failures are not alike. You already know that. And um, we human beings and our organizations make this worse. Have a natural aversion to failure. Like, we don't like them. We certainly don't want to be associated with them. We want to look good, not bad. We want to succeed, not fail. So, we have kind of an emotional aversion. We have a tendency to hide them, which of course has a direct connection to whether or not we can learn from them if they get hidden. Um, so, so part of, you know, part of my solution or answer to how do we help people learn from failure, put failure to good use, innovate effectively, and so forth, is to have a better classification scheme, right? A better, we will always have a natural aversion to failure, but I think we can overcome it when we have a thoughtful way to discern um, what kind of failure it is, what we can learn from it. So, I've identified three kinds of failure in my research um, basic, complex, and intelligent. I think the names give away that only one of them is really good, right? Only one is the kind of failure that we should all seek more of, not, not, not less of. Um, and indeed, we should do our best. And Ingred and I have both done a lot of work in in healthcare, and that's a context where it's a lot of innovation going on all the time. And a great deal of attention is paid, and rightly so, to preventing preventable failures. Both basic failures and complex failures are theoretically preventable, and when we're at our best, we prevent them, right? And then the intelligent failures um, are the ones we need to pursue.
So, just very briefly, a basic failure has a single cause, usually human error, but it's an undesired outcome in familiar territory where if we had been paying attention or didn't make a mistake, we would have gotten a different outcome. A complex failure, and healthcare is full of them, are multicausal. They're the kinds of failures that are sort of like system breakdowns. There were a handful of small deviations from best practice. Any one of which on its own would not lead to failure, end up causing a failure because of the way they came together. In healthcare, we often refer this to as a Swiss cheese uh model. And then intelligent failures are those that are still undesired. Um, if you're, you know, we, we want more of them, but if you're the scientist in the lab and you get one, you're still disappointed. Um, they are the undesired result of a thoughtful foray into new territory, an environment where we didn't yet know what would work. So, we're scientists, we're, we're inventors, we're innovators, or we're someone in our lives learning, you know, a, a new skill, and we're going into new territory, and we have to accept and even celebrate the fact that we'll get things wrong on the way to getting them right.
Do you have a favorite example of that? Of which? Of the intelligent failure? The intelligent failure. You know, in some ways, that's the hardest one because um, the easier, the others are easier to point to. Um, so I guess I have, I mean, I, I think the one I, I open the book with um, because I, I open the book with one of my own intelligent failures, and and it's in a research context, right? Which is in a research context, if you're not having intelligent failures, you're not a very good researcher, or you're not, you're not trying hard enough, right? You're not pushing the envelope hard enough. So, um, I mentioned earlier, I had a study. I was invited to be part of a larger study on adverse drug events or um, bad things that happened to patients caused by by error in their treatment. Uh, and I, I set out to essentially replicate research that had been done in aviation that said better teams um had, had better teams in the cockpit had lower error rates um than their than their counterparts, less good teams. So, I thought, okay, that that ought to be true here too, because patient care is enormously interdependent, and you've got a team, you've got to pay attention, good handoffs, all that, right? So, had a way of measuring teamwork, and trained medical experts were assessing the error rates essentially by going unit to unit and asking people what happened and so on. And for six months, I get the data, I run the numbers, and at first, I am just thrilled. There is a statistically significant correlation between my team data and the error data. So, I'm psyched. I just have to tell you, for those of you who are researchers or have any family members like that, that is the whole, that's the moment. Like, that's it. P value is all you're looking for in life. You're looking for the P value. I had the P value. P value. And then I looked more closely. I leaned into the screen and I realized, oh, it's in the wrong direction. So, yeah, I had that's not what you want. I had failed utterly. Right. And and, yeah, maybe it is. Right. So, I had failed completely to sort of get data consistent with my hypothesis. And of course, like any young researcher, um, I spent a couple of hours, you know, woe is me, pulling my hair out, thinking, what'll I do now? Because I'm going to drop out of graduate school. Um, and, you know, so like it, you know, it felt pretty painful. Um, and then, you know, I thought, well, a better thing to do is actually think like you have a failure, you try to figure it out, try to diagnose it. And so I started thinking like, what else might explain this result? I mean, it could be that bad teams make fewer mistakes. That doesn't make sense. Okay, what's going on? Anyway, long story short, um, it just occurred to me, you know, these are very threatening interpersonal events, you know, and I thought maybe the better teams aren't making more mistakes, maybe they're more willing to speak up about them, right? Like, that made sense all of a sudden. Of course, having that insight was a far cry from proving it. And I won't take you through the whole story, but um, it, that insight and the subsequent work I did to sort of see that that might be plausible led me down a very different path, and that was the path that ultimately became psychological safety research, and easily a more interesting thread of research than that which I'd been doing before. Right? So, the failure was a stepping stone to later success.
So, I'm going to keep going for the positive, but I, is there any time so we could sort of leave with the intuition that we all want intelligent failures? Like, that's the direction we want to go in. Yeah. Um, which I think makes complete sense. Is there? Well, I'm going to ask, um, are there any situations in which complex and basic are useful to us? I mean, they're such a part of our life, right? So, yes and no. I mean, they rarely bring new value, right? Or, or, you know, new knowledge that we didn't have before. Um, and they always, or almost always, bring new learning, right? So, so just because they're not helping us pave new territory doesn't mean they aren't tremendously valuable for learning, right? So, we can, you know, we can learn from, I mean, basic failures have the least learning, except you do always get to step back and say, you know, I wonder why I did that stupid thing, uh, that I should have known better, and it's usually, well, you know, I shouldn't, I actually have too much on my plate, or I was moving too quickly, and I know better, but I, you know, what's going on for me that I'm not actually present, not doing what I, you know, need to be doing in that moment to get the results that I'm perfectly capable of getting, or that our team is capable of setting. Um, um, but, you know, some, some basic failures are small, and some are enormous. Um, I document one, uh, plane crash that is, is technically a basic failure, single, single error, really. Um, and some are happening in, in, uh, you know, in US government just as we're speaking. I mean, just think, we're, we're in no, you know, we're, we have deep mature knowledge that that we could know in advance what might happen anyway. Um, so, um, but, but, but complex failures are rich sources of learning because um, they are the ones where if you bring a team together, and there usually is going to be a team, um, you can have tremendous aha moments where people realize that the way they were thinking about the situation was in conflict with the way someone else was thinking, and so they, they kind of crossed paths in in a way. So, when you take the time to debrief and and get the, all the blind men and the elephant together and say, well, what are you seeing? Oh, and, and what happened? And you discover errors of omission and errors of commission, and you can, you can redesign systems so that that kind of failure won't happen a second time.
So, there is across all three of them, maybe the, the common thread is if we take the opportunity to reflect on them. Exactly. Right. There's a reflection point that's actually regardless. You can't learn unless you reflect. Yes. And, and this is true even for intelligent failures, which are, you know, the most, most um, I guess the epitome of an intelligent failure is a hypothesis in a research study that was not supported by the data. That's an intelligent failure. But even there, it's often temp, okay, that didn't work. What do I do instead? Rather than pause, think, why didn't it work? What do we hope would happen? What happened instead? And what, let's see if we can figure out why. And I think that's just important for us to remember that the learning is a process in and of itself when we talk about learning, but you don't learn without the reflection that that is part of the process, and it's, it's disciplined, right? It's disciplined learning, and the, the actual deliberation about it. So, we, let's, you know, talking about failures, there are some, and I think you've already started to hint or say this, that help us move forward, drive growth, and there are other failures that help that we might say are hindering our progress, right? That's another way of carving the data, right? Do you want to sort of, does that, what I mean?
I, I do think I, you know, I gave the classification scheme of basic, complex, and intelligent. I think it's helpful. But you could also, you could also, if you wish, say sort of preventable failures and non-preventable failures. And the only pre, the only non-preventable failures are, are, are um, intelligent failures. And and yet they are avoidable. Right. Right. They're avoidable when we just don't take any risks. And it's probably, you know, it's probably more accurate instead of saying we should all try to have more intelligent failures, um, which is a little bit not actionable, like, let's all go have failures now, right? The real lesson is we need to take more smart risks. And I really do believe, you know, I believe that we should take risks, obviously, and I believe they should be as smart as possible, knowing that they will never be, you know, fully smart because we're not omniscient and, and, and we don't have a crystal ball. So, so I, I say the criteria to to say it's a smart risk, which is exactly the criteria to say it's an intelligent failure, is it's genuinely new territory, right? You can't just look up the recipe on the internet. And it's, um, in pursuit of a goal. You're not just messing about, throwing darts at the wall, right? It's in pursuit of a goal, u that you or others care about. You've done your homework, right? You've taken the time to find out, maybe from the literature, maybe from your colleagues, what do we know? What do we not know? And so that allows you to articulate a thoughtful hypothesis or to design a thoughtful experiment. And then fourth, and this is really important, no bigger than necessary, right? You don't want to, you know, bet everything you have on an uncertain outcome. You want to bet only as much as you're willing to lose, whether that's time, money, you know, reputation, um, or what have you. Be thoughtful about what you're willing, right? Yeah. Because, you know, you might feel very confident that this is going to work, but in fact, it might not. It's an experiment, right? And so realizing that failure can happen. Smart failures, and I mean smart experiments and less smart experiments.
So, up until now, we've really been talking about failure, but as you said, this comes from a positive side of being interested in learning. Probably many people in the room, long before they knew you for talking about failures, knew you for talking about psychological safety. And in my mind, and I think as you started to describe, they are intertwined. But maybe you could be very explicit about how these two things come together. So, there's learning, but how does, how do you know you're sort of known globally for this? And here's another thing, how do they, how do they intersect?
They're so connected, I can hardly separate them in my own mind. But here's, here's I think a good way to to um, talk about it. So, I said, you know, fundamentally, we have preventable failures and unpreventable failures. Well, psychological safety plays an invaluable role in both. So, psychological safety, I think first and foremost, and most of my early work is more on on this side over here, which is we need psychological safety so that people feel able to speak up, you know, quickly and honestly, so that we can catch and correct. I like to say that excellence today, it's not being perfect. Nobody's perfect, but it's, it's about catching and correcting the deviations that that do happen, right? So, when people feel like, wait, that's wrong, that's the wrong side, or that's, uh, you know, I made a mistake, um, or I disagree with you, that takes psychological safety. And, and those are incredibly important behaviors for preventing many basic failures and many complex failures. Um, on the other side, I believe in a, in a team or in an organizational context, you need psychological safety to be willing to experiment because you need to have that sense of fearlessness that I, um, I want to try something new. It might work, and it might not work, and that'll be okay. I won't be rejected or humiliated if this thing ends in failure. So, that also describes a psychologically safe climate. So, it's as important for, you know, error prevention and quality production, which you and I have studied together, um, as it is for innovation and risk-taking.
So, in that innovation and risk-taking, there's a lot of uncertainty, right? That's, that's the risk, right? If there isn't, if there, the uncertainty comes with the risk. How do we prepare ourselves for that? How to be, how do we be resilient in the face of insert, of uncertainty? And, you know, we're, we're living through uncertain times right now. So, this is actually a very important question as we think about whether we're living through intelligent failures or not. Like, how do we deal with this? How do we hold ourselves, ourselves? How do we bolster ourselves so that we can engage despite what is, I think, at least I will admit for me, maybe not for the rest of you, being uncomfortable with uncertainty, right? How do I, how do we, how do we do that?
Yeah, so we are uncomfortable with uncertainty, right? That's, I think that's part of the human wiring is to feel that sense of, we love c, we love certainty, right? I, I feel better when I have certainty, right? But, but given that I can't, then what? Right? So, I, I think it's really, I like to say, is there sort of three things we can do? And the first, and perhaps most important, is to, um, make it discussible, right? Call attention to it, like, sort of set the stage with, um, um, some thoughtful, and I think accurate, diagnosis of the context in which we find ourselves in. And by we, I mean, what, you know, the team, project, uh, the, you know, maybe we're in, maybe we're pilots flying to a, you know, ship full of people to somewhere else. Maybe we're nurses at the bedside. But being thoughtful about the context and referring to it often, right, which is talk about the degree of uncertainty. Talk about the stakes. I, I think very simply, and I think of context as essentially, how high are the stakes, you know? And if they're low, have fun playing around, you know, have fun experimenting. And if they're high, be very thoughtful, very cautious, and very much use the team that you have to avoid, you know, preventable failures. So, being thoughtful about what the stakes are, being thoughtful about the context, the, the uncertainty, calling attention to it gets us all on the same page to be clear about, oh yeah, this is what we're up against. May not be our favorite, but it is, and we're in it together, right? So, it's going to be okay. And the second thing, which is more straightforward, which is sort of create appropriate space and tools and skills for experimenting, you know, in, in smart, thoughtful ways. And then thirdly, um, institute rituals and responses to kind of, you know, celebrate the lessons that come from the undesired events that happen. You know, in, in sometimes in companies, they have failure parties, and that, you know, that doesn't mean breaking out the champagne, right? It, it means, you know, let's, let's come together and celebrate the hard work and figure out what we learned from it and just mark the occasion. But, but even more on a more microbehavioral side, I think we build resilience when when people are trying new things, they don't work, or when they're telling the truth, they're speaking up, they say, I need help, I have a different view, you say thank you, right? Thank you so much, you know, whether it's for that, uh, brave comment, or, you know, for that clear line of sight, or, or what have you.
So, let me maybe push a little bit on the to ask you to be concrete about creating space. Yeah. Yeah. What does that, what does that look like? Yeah. Right. Like, well, like if we're, you know, if we are, for example, you know, if we're, if we're right now in the moment trying to get these passengers to, there's no space there, right? Uh, there's space only for the best flawless execution we can do. But that also requires, you know, psychological safety, of course. Um, but be deliberate in carving out time and space for improvement or for innovation, right? And, and so whether it's, you know, in different contexts, this will mean different things, but be deliberate about it and be as structured as you can be. Right. So, the, the, the, the four criteria for judging retrospectively whether a failure is intelligent can also be used as a playbook going forward, like what, you know, in what new area might I like to make progress, um, and what, what's the goal I'm trying to pursue, and, um, what do we know about it? What more could I learn about it? And how do I keep the little risk here, because I'm uncomfortable with it, as small as possible, yet still be informative, you know, to me or to my team, right? So, giving people sort of structured, uh, tools and processes and skills to go forth into the uncertainty with less anxiety.
So, structure as a strategy for dealing with uncertainty. Exactly. Right. Which seems counterintuitive. It does, in a way. Structure is your friend. Structure is your friend, right? We think about, oh, we want to have flexibility, but at the same, same time, it's right, it's tricky. Um, a lot of us, and a lot of leaders have gone on to try to implement. So, imagine that you are leading an organization that wants to change towards embracing intelligent failures, right? And so, the question becomes, how do I do that, right? How do I? I feel like that's the question that people ask me the most about both about psychological safety and about this, right? So, how do I do this? That's sort of one question. I think maybe some of the space comments relate to that. But on the flip side of that, what do I need to be watching out for? Right? Where, where might I go wrong? And I, you know, sometimes when I'm teaching, I'm like, I make my students, Daniela knows, I'm like, okay, we're going to do the to-dos, and then on the board, I also want the team to come back with absolutely don't do this, or watch out for this. So, what, so do I do that while teaching? Are, are there?
I think so. So, you know, I, I think it's a, m, it starts with, um, shared meaning about the issues that we're talking about. So, I mean, I think we're trying to create a healthy failure culture, or a healthy innovation culture, or, you know, an organization that's more likely to have people who are willing to team up and do the hard things that uncertainty brings us and requires us to do. Um, you know, I think it starts with kind of calling attention to why this matters. And for instance, one of the, you know, one of the things I think everybody needs to understand is that it is a seriously, like, not failing is a seriously important failure, and it will look good and feel good today, but we will be in trouble tomorrow, right? So, making sure that message is out there early and often, right? That this is, this is why we have to keep stretching. We can't rest on our laurels. We have to keep innovating. We have to keep going, you know, bravely into the unknown. So, it's, it's, it's the dog that didn't bark, you know, we have to, we have to start listening. Yeah. If there's no failures, what are we doing, right? We're not experimenting. Something's wrong. We're taking safe bets. We're playing not to lose. So, doing that, um, you know, is again, sort of, um, depending on the context and so forth, giving people the skills they need. And, and I really, you know, I think it, a message comes from the top, but it's team-based fundamentally, like, do these behaviors are done with and around other people, with the people you work with. Exactly. And have a positive response to things that go wrong.
So, let me switch now, and please continue to send in your questions through the app. I'm going to start asking some of the questions that are on the app now. So, there's this first question. I think actually picks up nicely on what you were just saying, thinking about the team level. This person asked, "What role can should leaders, what, what role can leaders ask at different levels play in encouraging the opportunities for teams to experiment and take risks?" So, I think this is really a question about, you know, we can think about being at the CEO level, we can think about being at the middle manager level, and then there is sort of the informal team leaders. So, yeah, how do we think about what that looks like at every level? Is every level responsible for this?
Well, yes. Okay. Right. So, I think every level, or at least every level faces the opportunity to make a difference. I think I can't tell you how many organizations I've been in where I've, you know, I've met people who say, well, you know, they, Yes. Right. They aren't doing it, so we can't. And I say, nonsense, in the nicest way I can. Right. That in fact, um, all you can do is act and and influence where you are. And not only is that just true, but it matters. Like, you can literally create a pocket of excellence, learning, engagement, goodwill, warmth where you are. Like, your team can be a, a beacon of learning, and no reason not to do it. Yeah. And act, and there's good reason to do it, obviously, for it'll, the work will be helped by it, but even more importantly, it will be contagious, right? So, don't waste time and energy and heartache thinking about the things you cannot influence, but actually wake up in the morning and focus on the things you can. And even if you're not the team leader, the way you show up, if you show up as someone who is genuinely interested in what your colleagues have to say, and then you're one of those people who's sort of asking them questions like, what do you think about this? What are you hearing from customers? You're giving them the honor of being interested in their voice, right? So, it's a kind of, you're just making a difference. It might not feel like you're changing the world, but you're changing this world in small but palpable ways.
I'm really happy to hear you say that because I've said something similar, and I wasn't sure I was right, which is well, because I've long since graded you, right? That's fantastic. Yes. But I think this is a question that comes up for many students that I teach at are middle manager levels, like, I can't, not within my organization. So, how do I do it? So, I think that is really helpful. Right.
Here's another question. Um, can you suggest a healthy failure rate for an organization? No. Um, but I can, if you tell me more about it. Right. And, and, you know, if you think about a big organization, say, a big global organization, let's say a pharmaceutical or automotive, um, great, that's a good start. Now, you have to tell me which department it is, and or, or, you know, I, um, my husband's a scientist, and I, I asked him years ago because I was studying this stuff, you know, what, what, um, and I have to come clean, I had been telling people in talks that he failed, the experiments in his lab were failing 65% of the time. I just made that up. Um, and I, I finally thought, I should, I probably should ask him, right? That's right. You know, a little field research. So, I asked him, um, and he said, um, you know, he thought about it. He said, you know, I tried like to calc. I said, yes. And he looked at me like, you know, not a very supportive spouse, right? But, but, um, but it was, you know, it was a good answer. So, um, so if you are a basic scientist, a bench scientist, in a, you know, in a leading research center, 70, 80, 90 might be might be good, might be good rates. Um, if you're on the Toyota assembly line, um, I expect one in a million, 99.99% industry. Healthcare, right? And so, I, you know, I open chapter, um, one of, of, right, wrong, with the first cardiac surgery procedure on a live patient, um, in 1951. The patient died, right? And, and so you say, "Wait," and I say, "That's an intelligent failure." And you say, "Well, wait, wait, wait, wait. What about your fourth criterion? That that was big. That wasn't small." Well, how do you keep the, how do you keep it small when you're in new territory? No one's ever done this before. And the answer is, of course, that patient had no other option, right? No hope at all. Right? So, so in the early days of cardiac surgery, um, the failure rate was going to be extremely high indeed. We really couldn't figure out how to do this for a variety of complex reasons. Uh, today, um, the failure rate should be no higher than 2% in a major medical, major medical center. Um, and so it depends. It depends, but in, in some settings, um, there will be, you know, reasonable estimates you could come to. Oh, and you have to be hoping that a lower rate isn't because it's happening and they're not telling you. Yes. Yes. That then we're in real, we're in real trouble there.
Um, I think appropriate to this conference and the world we live in now, this question: How do you encourage intelligent failure with employees whose roles are ripe for Gen AI dis? I think it's going to be disruption. I think that's my guess. Like, what do you, right? So, these intelligent failures, particularly as we start to get more technological, could have some consequences, not if you, not in a good organization, right? And I realize most of them aren't, so forgive me. Um, but if you go with my definition of an intelligent failure, and it's, it's as small as possible. So, if you are an employee, and you are, I mean, one could argue, in fact, I heard Astro Teller at at Google's Moonshot Factory X argue just this. Um, if you're in a, in a certain kind of context, and you have no failures, you know, preferably intelligent ones, but no failures, um, then you, then you are ripe for layoff, right? Because you're not adding value, he would say. So, again, context depends. But if people understand, and I think it's important that they sort of, the message gets out and they understand what an intelligent failure is, um, then we're pretty proud of you. And given the enormous disruption and uncertainty and, you know, lack of, of knowledge at the moment about what AI is going to help us do, what it won't do, what, what it does well, what it doesn't do well. Um, we all should be experimenting, and if we're experimenting, we're failing. We're failing. But we want that to be in. But the anxiety underneath it, I mean, I get it. Like, I think we all get it. The anxiety about, um, what, why we're, whether we're going to be needed at all is very real. And I think that's at a much more global level, something we're all going to be wrestling with in the next few years.
So, I'm going to try and get in as many questions I can. Be a little concise. So, we have this question about, well, there are two questions here. Let me ask, what do we do for those who want to pursue intelligent failures, but they are already struggling?
Oh, well, you know, one, one, um, I think great suggestion is get a team, right? More often than not, smart risks are a team sport, right? Because they will be smarter if you've got other perspectives in on, let's try this. No, that won't work. I've already tried that. Oh, good. Okay, let's, let's tweak it. Right? So, then you won't be alone. And you should be faster, too. And you will be, you will be doing two things at once. You'll be doing good work with these smart failures, but also you will feel less struggling.
Here's another one that I think is interesting. So, this goes back to structures, but in a different way. So, this question, this audience member asks, how does intelligent failure perform in different organizational structures? So, if you were to think about organizations that are flat versus those that are hierarchical or have all many levels, is it different?
You know, it's reminds me of the research you and I did. Um, Ingred and I did a study some years ago of intensive care units. All of them had hierarchies. It was so, hierarchies a risk factor, right? Because people won't speak up. You know, there's sort of a, I need to look good in front of the boss, etc. Um, but what we found was enormous variance across groups, across hospitals and units in how hierarchy was handled. And that makes all the difference. So, hierarchy is fine. It's an organizing principle at times that works, but it's, do those who are higher in the hierarchy express curiosity, humility, interest in what others are saying and doing? Um, and then that creates a kind of learning environment that can allow us to be thoughtful innovators, experimenters, and so forth.
This person would like to hear from you. Can you identify an organization that has developed effective failing forward cultures? Is there a role model for you?
Yeah, I mean, there's sort of the, um, iconic role model, IDEO, um, which was doing this, you know, 25 years ago, um, and, you know, and trying to tell the world about it, uh, fail fast, fail often, right? But again, behind closed doors, safe, small, perfect, illustration of the intelligent failure concept. Um, and I think, um, uh, many, uh, many sort of scientific organizations I've been in, at least they have parts of them that that do this extremely well. Um, and I can think of others as well, but so there are some out there.
I see that I, so the thing that we were told is that we have to keep this session on time. That's how this conference runs. So, I see we have about 30 seconds left. All right. Um, so I think I'm going to open the floor to you. If there's anything that you sort of want to reflect for the audience, maybe lessons even as you have been talking about sort of sharing this work with people. Um, maybe if there's any surprising question that people have asked you along the way since you have been sharing this message of intelligent failures that has struck you or surprised you, and you've now reflected on it. Is there anything?
Well, now I only have seven seconds. So, uh, not your fault. Don't tell them the answer. Trust the question. Yeah. Yeah. Exactly. So, I mean, I think I, I have zero seconds. So, here's my favorite three words in the English language: Make it discussible. I think that's a perfect way to end. So, hopefully, you'll discuss with people while you're here for the conference. Everyone, can you join me in thanking Amy for being with us today?