707: Amy Edmondson on How to Build Thriving Teams with Psychological Safety

Pete Mockaitis
Amy, thanks for joining us here on How to be Awesome at Your Job.

Amy Edmondson
Thanks for having me.

Pete Mockaitis
I’m so excited to be speaking to you. You’ve been on our list for years, and so here we are. And so, I’m excited to dig into all of your wisdom, or as much as we can get, within the time we have available on psychological safety. But, first, I think we need to hear about you and competitive sailing. What’s the story here?

Amy Edmondson
How did that come up? I must’ve answered a question somewhere. Well, I was a competitive sailor as a child, not as much as a child can be, with my great friend Beth Haffner. We’d sail and race all summer and had a wonderful time. Then I sailed and raced in college, and then I took about 35 years off but started up again maybe six years ago. And it’s great fun.

Pete Mockaitis
Beautiful. So, now I’m curious, when it comes to competing, what is the nature of the event and the competition? And is there a team? What’s your style here?

Amy Edmondson
Yes. So, I compete only in the summer, in a small community in Maine where I’ve gone for many, many years. And I compete in a Sonar with two teammates, and there are only nine boats in the fleet, so that’s the limit to our competition. We race Wednesday nights and Saturday afternoons in July and August.

Pete Mockaitis
All right. And are things going pretty well, competitively speaking?

Amy Edmondson
Well, as a matter of fact, we just won the season.

Pete Mockaitis
Well-done.

Amy Edmondson
Yeah, thank you. It’s teamwork, it’s all about the teamwork and the psychological safety, of course.

Pete Mockaitis
All right. All right. Well, yes, let’s talk about psychological safety. First of all, well, I guess, whenever I hear your name, I think psychological safety, and vice versa. So, maybe first and foremost, can you give us your official definition? What do we mean when we say psychological safety?

Amy Edmondson
Well, recently, I’ve been thinking the best way to say this is just a sense of permission for candor. And the reason I say permission is that I don’t want to imply that psychological safety means it’s easy to speak up about, potentially, challenging issues, disagreements, or crazy ideas, or questions, or mistakes. But that there’s a belief that it’s feasible, expected, desirable, that people won’t think less well of you for it. So, permission for candor.

Pete Mockaitis
And I think that’s a great distinction because I’ve heard it said elsewhere, a definition of psychological safety is the belief that you are able to say whatever is on your mind without fear of a negative reaction. And I thought, “Hmm, I don’t have that relationship with almost anybody.”

Amy Edmondson
Right. At least without fear of being marginalized or penalized in some way. We all are human and we will have negative reactions to disagreement or certain kinds of bad news. It’s just our emotions will kick in quite quickly. But if we’re thoughtful and we’re a good team and we’re committed to doing the best we can, we will catch ourselves, and say, “Don’t shoot the messenger.” Just because you just said something unwelcomed doesn’t mean I should shun you or think less well of you.

Pete Mockaitis
That’s good. So, not marginalized, not penalized, not shunned, not thought less well of. But naturally, someone might say something and you think that they are mistaken, and you maybe even feel underappreciated that they would bring up such a thing, but you’re not going to, like, punish them over that even if you have a difficult interaction.

Amy Edmondson
Right. And I think that’s easier to do if you have an honest appreciation of what you’re up against, meaning the nature of the work requires stumbles and falls along the way. If you’re talking about doing something that’s utterly routine and well-understood and well-known, then maybe your expectation should be of only perfect comments and only perfect performance.

But if you’re doing work, like most of us are, where there’s lots of potential for wrong turns and screw ups along the way to greatness, then that’s just part and parcel of what we’re doing here. So, it helps to have a clear-eyed sense of what we’re up against and what we’re doing.

Pete Mockaitis
All right. And then, tell us, that sounds like a pleasant thing, “Oh, yeah, psychological safety, I’d like to have that with my friends, family, colleagues, collaborators,” but more than just a sort of nice to have and pleasant vibe, psychological safety has huge implications for performance. Can you share with us a little bit about that relationship and some of the most compelling bits of data or stories?

Amy Edmondson
Sure. It’s funny because the variable, the measure of psychological safety that I’ve developed 20 years ago, it’s been around in the research literature for a long time, but it’s now been more widely used in company settings and so, in other words, we have more and more data on some of the benefits of psychological safety.

Probably, the most visible, widely read that they’ve done at Google called Project Aristotle, and that was about five years ago. And the study set out, it didn’t set out to study psychological safety, it set out to try to figure out what are the key factors associated with differences in teams at Google, so they studied 180 teams. It turns out they tested about 250 different variables, and psychological safety emerged as the number one predictor of performance in teams, so the number one sort of explainer of variance across teams.

And I think it’s a strong statement to say that surprised them because if you’re looking for something, it’s easier to find it. If you’re not looking for it, it’s almost a more compelling discovery when it pops up as the factor that really helps explain these differences. One of the things I like about that study, too, is that it shows very clearly that there were differences, differences in performance and differences in effectiveness across teams at Google.

So, it helps us see that this is something that varies across teams even in the same corporate culture, and that’s important because we then can be very clear about the fact that psychological safety isn’t just mirroring the culture. It’s climate. It’s interpersonal climate. And even in a very strong or very interesting or healthy corporate culture, you can still have differences in interpersonal climate, differences in just subtle willingness to be candid, to speak up, or to not hold back.

Sometimes I think it’s easier to explain that the absence of psychological safety is basically a preference for, “Oh, I’ll just wait and see. I’ll hold back and maybe things will clarify, and then maybe I’ll speak up.” But that’s an awful lot of cognitive work. So, putting that aside, so the Google study is a good study of the nice relationship between psychological safety and performance, and many others.

One of my favorite studies that I did, which was in a healthcare delivery setting in intensive care unit setting, 23 North American hospitals, 23 intensive care units, we found a statistically, significant relationship between psychological safety and quality improvement. So, the ability over time for teams to improve the quality of care, which was ultimately associated with lower rates of morbidity and mortality, that’s harm and death, so that’s a pretty strong one where life and death are concerned. There are many others though. They’re now really hundreds of studies that have relationships to things like performance, learning behavior, quality improvement, you name it.

Pete Mockaitis
Okay. Well, that’s a nice overview. Thank you. Could you now perhaps paint a picture of kind of across, I know I’m sure it’ll vary greatly, well, hey, even relationship by relationship, let alone team by team or workplace by workplace. But kind of, roughly speaking, what’s the median average-ish level of psychological safety in workplaces today in the US? And I don’t know if you want to give me a number or paint a picture for kind of like the theme or the vibe.

Amy Edmondson
Yeah, I’ll paint a picture.

Pete Mockaitis
What’s the typical psychological safety story in a workplace these days?

Amy Edmondson
Well, it’s probably a fool’s errand to try to say what’s typical because there’s so much variability. And even during this difficult time of COVID, there’s been extraordinary variability in terms of some places. I think that the rallying together to do what people can to sort of make things work during these difficult times has created the stronger sense of a bond and more psychological safety where people realize, “Yeah, it’s okay for me to say what I’m thinking and to get help when I need help, and that’s acceptable now.”

But in other places, I think where people, especially in workplaces where people are being asked to do things they might not be comfortable doing, one could arguably say that psychological safety has gone down. I strongly believe that, in most organizations, there’s still variance across groups. And this is, in part, because psychological safety is a very local thing that this team might have and that team doesn’t, and that may mean, this is really a middle manager thing, or team leader effect more than, say, a CEO effect, and that’s very much been the case in all of the datasets that I’m aware of.

But, still, I’m dodging your question, saying it depends, there’s lots of variance, some people have it better than others. And, yet, there’s no question in my mind that, nowadays and even before the pandemic, it’s not high enough. So, I think it’s fair to say that very few workplaces have as much psychological safety as would be optimal in terms of helping people do their very best work and helping people team up effectively and solve problems.

Fortunately, the average workplace, I’d say, is not one that’s incredibly toxic or incredibly fearful where there’s a complete focus on self-protection as opposed to on the mission or on what our colleagues need from us and, really, a state of fear. I think that it’s out there, for sure, but it’s not the dominant workplace.

And then I would say there’s few where it’s just extraordinarily high where people are candid and aware of their fallibility but ambitious about what they might do together, and they sort of engage in dissenting views and conflict and problem-solving without fear of reprisal. That’s the other end of the spectrum. In the middle is a whole range of places where, in fact, it’s not toxic, it’s not terrible, but, on average, there’s still too much holding back. People are holding back their ideas, their perspectives, trying to look good in front of their colleagues, their managers, and it limits their ability to contribute.

Pete Mockaitis
Okay. So, the “average” or “typical,” which is hard to do, looks like not just straight up abusive hardcore toxicity and fear rampant, but plenty of people holding back in order to look good and concerned about speaking up, and could be harmful or problematic to them. So, I guess I’m curious, if we think about making the leap from kind of “average” or “typical” the suboptimal picture that most of us find ourselves into, versus approaching best in class, well, maybe could you give us a cool case study of do we have a transformation there in terms of what was it like before and what was the vibe like after? And then how did that translate into some results?

Amy Edmondson
So, one of the great turnaround stories, and I do write about this in some detail in The Fearless Organization is Cynthia Carroll, CEO of Anglo-American, which is a mining company in South Africa. And when she became CEO, which is already a stunning thing because the first woman CEO and so forth, she was appalled to discover the degree of worker accidents and even deaths.

And so, she decided to make that her mission to profoundly transform the performance on this crucial dimension of workplace safety. And to do this, she realized pretty quickly that she needed people to be speaking up, speaking up about unsafe conditions, speaking up when they’re being asked to do something that’s unsafe, or when they’re sort of aware of a hazard.

Not easy to do because it’s been decades, even generations, of not being heard and not being listened to and feeling that you just go in there, you do your job, and that’s that. It was a pretty stunning kind of intervention, got everybody in the stadium and got them talking in a new way, and was able to kind of apply that into the workforce and turn this around and make a dramatic difference.

Here’s a very different context. SED, one of the largest Nordic banks, did a sort of, I wouldn’t call it as a dramatic turnaround because I don’t think they weren’t in real trouble, but senior leaders were aware that the financial services industry was changing, more fintech players, more innovative. And the executive who ran the risk group decided, that psychological safety for speaking up about potential risks.

Because when people just feel like, “Ooh, maybe I’m wrong,” and they hold back and they’re not confident enough that their superiors want to listen to them, the bank is more vulnerable to risks. And so, that was a very thoughtful turnaround of that unit, and then it started to spread to other business units in the bank as well. So, that was a fun one to write about.

Pete Mockaitis
Indeed. All right. So, that’s sort of the picture there. And I’d love it, in terms of sort of the practical how-to, if folks are in organizations and they want to improve the psychological safety for themselves and others in their teams, what are some great starting points or key practices that make all the differences?

Amy Edmondson
A way to answer that question, in terms of both as a starting point and a practice that makes a difference, is start with the work, start with how the performance goals that you share, what they look like and what they require so that we’re not doing this just for the sake of doing it, or because we’re interested in culture change per se, but we articulate sort of why the work we do needs us to behave and show up in a different way.

So, articulating goals that matter, that are motivating, that are energizing, and then kind of having some discussion about why achieving those goals requires people to voice their ideas, to challenge each other, to be open about failures, is sort of the next logical step. And then I think it’s really important not to dictate how we’re doing to do this but to invite people to sort of suggest some things that they think might work, that might help them have an easier time offering their ideas or asking questions. And then start testing some of those suggestions, and just keeping it in the context.

I’m advocating not for, “Let’s go offline and learn some things,” but, “Let’s practice some new ways of talking and being while doing our work.”

Pete Mockaitis
Okay. That sounds good. So, that’s a nice thing to kick it off and kind of get folks engaged, rallied around that goal, and it makes sense. It’s not some extra thing, but it really has impact on what we’re trying to do here. That’s cool. And then I’m curious about just sort of like the basic ways in which we talk to and interact with each other. Like, what are some dos and don’ts in terms of kind of offering feedback, asking for input, responding to failure? I think some of us might need a pretty dramatic re-programming of just the way we talk to people.

Amy Edmondson
That’s a great way to put it and it’s hard. I struggle with this question. It’s not the first time I’ve thought about it. But I struggle because there’s no easy answer. It’s, “How do you do re-programming?” And I talk a lot, I think a lot about framing and I talk about reframing. So, framing is something we do all the time as human beings. We think we’re sort of under the illusion that we’re seeing reality. We’re not. We’re seeing reality filtered through our own beliefs and all the rest.

And sometimes our frames are really obsolete. There are frames that we inherited from an earlier era, an era when the relationship between kind of effort and results was more straightforward. You tried really hard; you’d get the results because the formula was pretty clear. Follow the recipe, you get the results.

And as an increasing portion of the work doesn’t really conform to that simple frame, we have to explicitly and deliberately reframe which is another way of saying re-reprogram to help ourselves really appreciate that we’re fallible human beings in a complex uncertain interconnected world. Those are conditions that will necessarily give rise to the unexpected and the undesired and, also, some, now and then, happier surprises.

So, that re-programming, in a way, it helps us get over ourselves. We’ve got to shed the idea that we need to be perfect. We’ve got to shed the idea that we need to look good all the time. And I know, I suspect most listeners don’t think, when they say, well, I’m not telling I need to be perfect or I need to look good all the time, but, in subtle ways, we’re acting as if that’s the case. We’re holding back too often. We’re putting the threshold for when we should speak up higher than it needs to be.

And so, to do this re-programming, I think it’s a lot of having a kind of cheerful recognition that you’re a fallible human being in a fast-paced uncertain ambiguous world, and then, “Ooh, if I really appreciated that that was the case, how would I show up? Well, I’d ask a lot more questions. I’d be a lot more curious.”

So, the re-programming starts with that kind of clear-eyed acceptance of reality and realizing that might be different than how we kind of tacitly think about reality. And then forcing ourselves to be curious, which then allows us to do what I think is the most important thing of all, which is to ask more questions, genuine questions, like you’re doing. You’re asking me questions, and then you are quietly listening to the answers. If only real life were like this, not just podcast life.

Pete Mockaitis
Right. Yeah, that would be nice. And so then, that’s a great frame for starters in terms of, well, boy, say it again. I’m a fallible human in a changing…that’s so good. Let’s hear it again.

Amy Edmondson
Okay. And I might not say it the same way twice, but I’m a fallible human being living in a fast-changing uncertain interdependent world.

Pete Mockaitis
Yeah. You know, tell you what, just sitting with that, for me, in this moment, is just bringing a sigh of relief, you know, in terms of like I can let go of a lot of pressure, stress, expectation that need not be there.

Amy Edmondson
Right. I’m the same way. I talk about this, but do I practice it consistently? No. In fact, I have this to-do list that I started with this morning. It’s utterly unrealistic. There’s no way I can, you know, get, “Oh, I’ll finish a chapter, I’ll have this wonderful time with you.” It’s crazy. But I do it every day as if. And then I feel bad about not getting through it.

Pete Mockaitis
Yeah, that’s great. And then if you really do internalize that conviction, it’s like if someone…even though someone does kind of disrespect you with regard to, it’s like, “Really, Pete, that’s on page four of the briefing document. Like, that was a really stupid question and I’m appalled that you asked,” in that tone of voice, face, which is where I think about this, the violations of psychological safety left and right.

You can feel better about that, it’s like, “Okay, yeah. Well, yeah, fair enough. I should’ve read the briefing document before making…that’s true. Easy mistake I made,” but that doesn’t mean I’m bad or a loser or worthless, a team member who doesn’t belong here. It doesn’t mean any of those things. It’s just like, “Yeah, I made a mistake. We all do it. Yeah, moving on.”

Amy Edmondson
Right. And I’m not a fan of making the same silly mistake multiple times in a row. We do have to learn from and keep striving to do better, but I imagine most people feel that way as well.

Pete Mockaitis
And so then, I would love it if there are any particular words or phrases that you see and love in psychologically safe organizations versus see and really irk you in not so psychologically safe organizations because I think there’s just a lot of little subtle ways that psychological safety is built and destroyed. Just for example, one of mine is when someone says “Obviously,” I really don’t like that because it’s like, “Oh, I didn’t know that I must be an idiot.” That’s one of my pet peeves.

Amy Edmondson
That’s a beautiful example.

Pete Mockaitis
And I think most people need not say the word obviously in most of their business communications, but that’s just sort of my hobby horse.

Amy Edmondson
Absolutely.

Pete Mockaitis
You tell me, Amy. What are some of yours?

Amy Edmondson
No, that’s a really good one. And, because with compassion, it can be a habit. So, it’s a very counterproductive word to use in interpersonal communication for the reasons you just articulated. And I’m aware that I accidentally do use it sometimes because my brain speaks that way to me, and then I use it. So, that’s okay as long as we can sort of catch and correct and occasionally laugh at ourselves for doing that. And I sometimes will, I’ll use the word, like obviously, and then I’ll stop and say, “Oh, no. So, if it were obvious, I wouldn’t say it,” or it wouldn’t be a nice way to say it anyway.

Another one is “To be honest.” I mean, crazy to say that because it basically invalidates so much of the prior conversation we might have. So, if I say “To be honest,” it’s like, “Wait a minute. Was everything up until now not really honest?” And so, these kinds of things can be well-meaning but problematic. It’s such a good question that you just asked that I’m going to now commit to creating a list.

Pete Mockaitis
Oh, please let us know and we’ll link it in the show notes and share it if we can when the time comes.

Amy Edmondson
Perfect. That’s a good idea. That’s a good idea because I do love studying conversations, studying the actual exchange of words, and noting those problematic triggers that sort of indicate, any word that indicate, “Oh, you’re supposed to have known that already,” or, “Your question isn’t really welcome,” you name it.

Pete Mockaitis
That’s true. Yeah, those are good categories right there, like, “Your question isn’t really welcome.” We think that, “You’re dumb. I think you’re dumb because you said that.” I remember once, I was working on a project in retail, and, again, it’s these little things. And so, it was a major department store, this was a consulting project, it was a major department store, and we were learning about size packs, which was a new concept to me in terms of, like, if you buy it from a clothing designer, I don’t even know if this still works this way, but you can choose from size packs.

So, a size pack might have four extra larges, ten larges, three mediums. And that was really surprising to me, and I was like, “Wait a minute. So, we’re a huge department store client, right? And we got these clothing suppliers…?”

Amy Edmondson
Pretty limited, yeah.

Pete Mockaitis
“We can just tell them, no, I want exactly these many smalls, mediums, and larges. Like, really, size packs? Like, why do we do that?” And I remember the partner on the case looked at me, and he said, “Are you serious?” Like, he genuinely didn’t know if I was trying to make a joke, but I really wasn’t. But when he said that, I was like, “Oh, apparently, that was a phenomenally stupid thing to say.” And I still don’t know why to this day, I’m like, “If you’ve got the market power, shouldn’t your suppliers give you what you want?” I don’t know, but maybe there’s a logistical supplier reason and trucks or packaging or something less known.

Amy Edmondson
Well, it’s easier for them, clearly. But, “Are you serious? Because, as you said, “Are you serious?” as a sarcastic statement, which it may have been, is problematic. But if it were genuine, I’m in favor, “I just need to check, I’m not sure. Are you serious or are you…?” so, anyway.

Pete Mockaitis
I hear you, yeah.

Amy Edmondson
The genuineness really matters.

Pete Mockaitis
It was genuine. And then I think that’s another layer to this psychological safety stuff. It’s like you could be speaking perfectly safely, and someone could still receive it negatively.
So, for example, that partner said, “Are you serious?” and even if it was genuine, he was like, “Are you serious?” And I really was, but the fact that he sounded serious made me think, “Oh, apparently, this is so obvious I’m a moron.”

Amy Edmondson
Yeah, right, that’s true. That’s true. And then you backed down.

Pete Mockaitis
That’s the conclusion I leapt to but that’s on me. He didn’t demean me or he wasn’t rude to me.

Amy Edmondson
Yes, that’s great. That’s on you. That’s you withdrawing and feeling, “Oops, just slightly less safe,” expressing your thoughts about this work-related matter, even though technically it wasn’t his fault because you put sort of that embarrassment on yourself, you said, “Oh, I guess this is something I’m supposed to know. And maybe I stepped out, I tiptoed out, and it didn’t work out well, so now I’m going back into my shelf.”

Pete Mockaitis
Yeah, that happened. And I guess I’m curious, given that human beings with our varied triggers and hot buttons and sensitivities, that can happen, any pro tips for dealing with that and trying to continue building psychological safety given that reality?

Amy Edmondson
Yes. Interpersonal skills are skills that we can continue to develop our whole lives. I don’t think anyone ever perfects them. And the interpersonal skill that I’m deeply interested in, because of its relationship to mutual learning, is that ability to kind of have an honest conversation, especially about a misunderstanding, like in that moment. Now, don’t think you want to do a deep dive in every crosswire that might happen throughout the workday, but, occasionally, that one really stuck with you, that really struck you.

Pete Mockaitis
This was a decade ago, yeah.

Amy Edmondson
You were puzzled by it. It stuck with you. And so, occasionally, really, it’s worth saying, “Hold on, could we do a quick timeout here?” or maybe if we’re too busy now, “I’d love to talk about this later. I need to understand better. Here’s how I was seeing it. Am I really missing a sort of area of expertise in this industry that I need to develop? Or, might this possibly be an area of innovation that we could work on together?” And so, that’s the substance.

And then the interpersonal substance is, “I felt bad and maybe even assumed that my ignorance was glaring in that moment when you said that, but I understand why you said it.” So, that we can sort of start to develop working relationships with people where we understand each other’s needs better, and then we’re better able to learn together.

Pete Mockaitis
Yeah, that is good. And, as I imagine, I’m sure there’s all kinds of potential sensible explanations under the surface, like, “Oh, I’ve been working in this industry for 20 years, so size packs are just like second nature to me.” But, yeah, yet you might think that…whatever. So, I could see how that unfolds. And then, over time, certainly, that feels great in terms of relationships being strengthened by engaging in these exchanges. All right. Well, then could you tell us a favorite quote, something you find inspiring?

Amy Edmondson
“I haven’t failed. I’ve just found 10,000 ways that didn’t work,” and that’s, of course, Thomas Edison. And it’s this notion that in new territory, which all of us are in, more and more frequently nowadays, we reframe. We have to reframe how we’re thinking about the things that go wrong so that we actually understand them as progress toward the things that are going to go right. So, that’s one in terms of the substance and just sort of feeling better about ourselves when things don’t go the way we had hoped.

The other one is a quote from Abraham Lincoln that I adore because it speaks to this interpersonal realm. And he said, “I don’t like that man very much. I must get to know him better.” To me, that’s a very profound statement. Most of us, “I decide I don’t like someone. I’m going to, okay, I don’t like him. I’m going to go spend time with other people.” It doesn’t occur to me, instantaneously, to think, “I don’t like him. I guess I don’t understand him well yet. If I understood where he’s coming from and what he cares about and his hopes and dreams, I’d like him.”

Pete Mockaitis
And could you share a favorite study or experiment or bit of research?

Amy Edmondson
I’ll have to say that a favorite study was the study that didn’t turn out the way I wanted it to, was the first real project I did as a graduate student, as a PhD student, where I was trying to show the better teams in a healthcare delivery setting had fewer errors, and the data, once I had it and analyzed it, seem to suggest the opposite. In other words, the better teams, according to the team survey instrument, had higher not lower error rates, like, “What? What’s going on?”

Well, that was the surprise, undesired result that led, ultimately, to you and I having this conversation today because I was able to figure out that, right away, with a lot of extra work, that the reason for this result was that the better teams were more open, more honest, more willing to report error, and so it looked like they had the worst error rates. But, in fact, we don’t know the denominator, we don’t know what the real error rate was for any of those teams, but we did find out, ultimately, they had very different interpersonal climates, which I would then call psychological safety.

Pete Mockaitis
And is there a particular nugget you share that really seems to connect and resonate with folks; they quote back to you often?

Amy Edmondson
Leader is a position, leadership is an activity. Anyone can exercise leadership.

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

Amy Edmondson
AmyCEdmondson.com or Harvard Business School Faculty page, HBS.edu. Go to Amy Edmondson there.

Pete Mockaitis
All right. And do you have a final challenge for folks looking to be awesome at their jobs?

Amy Edmondson
Ask more questions.

Pete Mockaitis
All right. Amy, this has been a treat. Thank you and I wish you many fun adventures in sailing and more.

Amy Edmondson
Thank you. It’s been a treat talking with you.