How to Thrive in a World of Continuous Transformation with Phil Le-Brun and Jana Werner, Executives in Residence at AWS

have to go incredibly fast in order to get anywhere. So why are organizations not set up right to take advantage of this, 

Jana Werner: Phil? 

Phil Le-Brun: They don't need to go incredibly fast, they need to go less slow.

It's y- much of our structure has been deliberately put in place over the years 'cause the cost of execution, the cost of doing things was really expensive. So when you did something, you wanted to make sure it was right. And often, leaders and the leaders of leaders knew what that end-to-end picture looked like more than the people at the front line.

So you had all these decision-making governance processes, but it made sense because you wanted to get the answer right. Now, we're in an environment where I can run an experiment on my laptop in five minutes, yet it takes six months to actually get approval to do that. So our organizational decision-making processes haven't kept pace with the rapid evolution of the technology.

So what's holding organizations back is simply this idea of being able to experiment faster as actually an offset to risk. 'Cause, often you think about if we move fast, there's more risk. It's the opposite. The fact it takes us so long to make a simple decision is a bigger risk to businesses they suppose they'd supposed in the past.

Richie Cotton: When things are expensive to do, the cost of getting things wrong is very high, and so that seems to have fundamentally changed. So what needs to change with organizations then in order to deal with this new world? 

Jana Werner: They need to work away from this idea what in our book we call Tin Man organizations, this rusting construct where we have inherited Frederick Taylor and co's way of operating.

It's top-down, built on standardization, specialization, control, individual performance, compliance. That all worked when you had predictable outcomes and... but it's all built on a foundation of permission- To innovate, even to speak up. And of course, that works to de-risk your organization with predictable outcomes, but we're not in that place anymore.

So we talk about how do you get less transactional, more cross-disciplined, and we see a bunch. We've... I think we talk about 36 anti-patterns that need to change. You push ownership into the organization more. Don't take decisions that have to travel up to the top back down. You get out of the silos. You do small cross-discipline teams.

And around that a lot needs to happen to re-inspire curiosity, and for leaders to create the context and the clarity that people can operate like that and be really fast in small teams. 

Phil Le-Brun: Leaders get this. We sit in rooms with leaders, and they'll flick through the book and read an anti-pattern, and they'll start to laugh 'cause they can s- it's y- We do the same things as everyone else because it's safe.

And then when you actually stand back and look in the mirror and say, "Why do we do-" To have 42 governance committees. Why do we hire bright, intelligent people and then put 12 layers of management on top of them to tell them how to do the job we hired them to do in the first place? It's... when you start to storytelling it and hold a mirror up, it's obvious that what we do today isn't really suitable for the environment we're operating in.

And most leaders get that. The question is, how do you actually then change? 

Richie Cotton: Yeah, so I love this idea of anti-patterns, where things go wrong. It's almost every workplace you're in, there's something everyone kind of knows is crazy, but somehow you've not managed, able to articulate what's going wrong.

So yeah, we're definitely gonna get into some of these anti-patterns from your book in depth. But Jana, you mentioned the idea that a lot of organizations are like Tin the Tin Man from The Wizard of Oz. And for anyone who's watching the video version of this, you can see behind Phil is The Octopus Organization book.

Why did you pick the octopus as a, a metaphor for what a good organization looks like? 

Jana Werner: An octopus is an absurdly sophisticated animal. It's able to adapt through shape-shifting. It can change its texture, its skin. It even can change its RNA, so its chemical makeup, within, very fast, within a few hours, to go from hot, for example, to living in colder waters.

It changes within hours, and this is really amazing. We thought this animal is so curious. It learns really fast. It has to grow up without parents in the ocean. But most importantly, it has two-thirds of its arms in its of its neurons in its arms. So the arms can can s- sense, adapt, and react separately to the brain, but also together.

And because there are no perfect organizations, we didn't want to name organizations. We picked a metaphor that describes of how you have to push intelligence to those close to the customer, to the technology and with a passion to drive change. 

Richie Cotton: Okay, yeah, I like that. You got your brains in your arms or in your tentacles, the people who are actually doing the work are the ones who are more empowered there. You're making things a bit more distributed. Actually, Phil, did you want to add into that? Are there any more features of octopuses that you like? 

Phil Le-Brun: The fact it doesn't have a hard shell, so the f- its defense is its agility.

It doesn't grow up with parents, so it has to learn its way very quickly around through the world. But it's... The metaphor I've only just thought of this. If you think about organizations, they change their structure quite often, but they take forever to change their culture and how to make decisions.

Octopuses are the opposite. They haven't changed their structure for hundreds of millions of years, but as Jana said, they can adapt their RNA very quickly. So it's almost like organizations need to invert how they think about themselves. 

Richie Cotton: Okay. Yeah sounds like a very big change for a lot of organizations.

I think yeah, not many organiz- Oh, mind you, GitHub had the octopus as a logo. May, may, maybe GitHub is an octopus organization as- But, Yeah. Don't know whether that's true. We can maybe test it. But I'd love to get into some of the anti-patterns that are in your book. So there were a few of them around data, so I wanna start with those.

And there was one about misusing metrics, which I think i- is a very common one. So talk me through what's a bad use of a metric? Yeah Jana, do you wanna go first? 

Jana Werner: Yeah. The, our current favorite one, because we advise on AI transformation a lot right now, so our current favorite bad metric is when people ask about AI adoption metrics.

This is always coming back to the same trap. What gets measured gets managed. And people will check how many people are using AI, but that doesn't tell you anything about whether they use it for something that has high quality, that creates really high productivity. Asking the same simple prompt that gives you the same simple answer is certainly not the same thing as someone creating a complex agentic network that is able to create a crazy productivity improvement.

We see a lot of leaders ado- obsessed with AI adoption. We're telling them, that doesn't, that really doesn't matter anymore how you do that. It drives the wrong behavior. We loved... We love Goodhart's law. He was an, a British economist. He says, "Once a metric becomes a target, it stops being a true measure," and people will destroy the company to get to the metric they're measured by.

And so we try to help leaders understand what the properties are of good metrics. 

Richie Cotton: Yeah, it's very true. So earlier this year we had this craze of token maxing, where every organization was trying to spend as much as possible on AI, and so yeah and although that is completely dif- divorced from are you actually adding any value?

Okay I love these examples. Phil, do you have more examples of terrible metrics? 

Phil Le-Brun: There's all sorts. There's vanity metrics, like how many times has my application been downloaded? It's just, who cares? How many people are actually using it? There's perverse metrics, which don't have the impact you expect.

There's a great story where in India, the government used to reward people for killing cobras. Guess what happened? A whole cobra breeding industry took off in the country. So it's really about thinking about what the metrics you're putting in place are for, and we think of metrics less about a target, more like a vital sign a doctor uses to help you live well, not a number to hit.

So something you can learn from and adapt. 

Richie Cotton: Okay, yeah. I... The cobra example's completely crazy, right? You're trying to get rid of cobras, and then people start breeding cobras to get rewarded, and you get more cobras. So it has a, the opposite e- effect of what you intended. So how do you create a metric that isn't gonna be gamed in that way?

Like, how do you create a good metric? 

Jana Werner: I think it's more about the intent of metrics. That's what we're trying to help leaders understand. If you use a metric, the point usually shouldn't be so much whether a team hits the metric that you give them. It's usually should be an intent to learn. That's how I...

We both work at Amazon. That's what I've learned at Amazon. We use metrics to learn, not to hit targets. It's a bit like vitals that a doctor uses. It helps you to live better. The idea isn't to hit the perfect target, but you want to learn how to get better in your body. So the other thing we think about is favoring input metrics.

It's not like looking in the rear view mirror when you measure something that's already passed. But what are the inputs that you can control to get to better outcomes? Just give you one example. In, in the early days of Amazon Prime, the team measured in-stock items that you could order and so that you could, have them there fast to really order them quickly.

And it... But they linked it back to what mattered. Could customers get the right things when they wanted them? So good metrics can earn their keep, and you keep using them because they make sense. For others, you better ask, "So what?" three times. If you can't connect your metric to a decision by the third "so what?",

then just get rid of it. 

Phil Le-Brun: Here's the thing. Every metric can be gamed if someone so chooses to. So the way we look at it this is it's about changing mindsets around the metrics. It's not by adding even more inspection to them. So when people fear the numbers or fear the metrics you've given them, you'll get what we call the watermelon reporting.

Everything's green until the last minute, when it's red, just like a watermelon, green on the outside, red on the inside. And one team we spoke to did exactly that, and it took a new leader six months to show that there was support for them. They weren't gonna be punished for not hitting a metric. That metric was there to learn, and only then did the team start reporting the right metric.

So it's given them the comfort the rationale, the reason why that metric exists, rather than just blindly giving them a metric to try and hit. And a lot of this starts with education of leaders. We had a Amazon leader who jumped at the AI adoption metric, for instance, and we explained to her why that is futile, why adoption, as we talked about, means nothing.

It's, how do you achieve value, and how do you measure things which- Lead to delivering value with AI. Yeah, having 100,000 people use AI today is a meaningless metric. Having 50 people use it to deliver value, now that's something interesting. 

Richie Cotton: Absolutely. I love the idea of just thinking about why do you have these metrics in the first place rather than just everyone following these things.

I'm curious what your take is of having targets that are tied to remuneration, like sales teams are notorious for this. Like you get bonuses if you hit specific targets. I guess the same to a certain extent with executives. How does that affect things? 

Phil Le-Brun: It goes back to what Jana was saying. If if you give people just a pure metric to hit, people will do everything they can, even if it destroys their company.

And that sounds quite extreme, but if you look at how many companies are run today, the- they are measured on what did I achieve this quarter? What are you gonna do? You're gonna cut costs, even if it sacrifices your future, to hit your quarterly bonus. So we talk in the book about the difference between incentives and rewards.

Incentives are a bit like the carrot and stick. It's a bit like saying to an executive, "If you hit your quarterly number, you get a bonus." Guess what? They're gonna hit that number regardless of the consequence. Rewards are something more intrinsic, and it sounds a little soft but, at the end of the day, we truly believe from the people we meet, they want to do the right thing.

And simple things like calling out good behavior, people who've experimented, whether it's a success or failure, that has this massive impact on people 'cause they actually wanna feel like they're doing a good job, doing something which has a bigger purpose than just hitting a short-term metric. 

Richie Cotton: I love that idea of just sharing wins or even like talking about, like someone's done something well even if the project wasn't a success.

Just yeah calling out like, "This is a good thing. All you other colleagues, all you other employees should be doing something similar." Yeah I can see how that can have a big impact on cultural change. The next anti-pattern I wanted to talk about, I think this might be a bit sacrilegious to our audience, but this is about deferring to data where you let the data make your decisions for you.

What's wrong with data-driven decision-making? 

Phil Le-Brun: You can get blind to what's really happening in the world. What's wrong is when data-driven quietly becomes data is the final word. We call this anti-pattern deferring to data, and your data does feel objective, doesn't it? It's a number. You can hit a number.

You can look at a number and say, "This is fact." But it's usually an incomplete picture, and it's open to interpretation as we talked about with measures, and it's one view of a complex system. And when it becomes the thing that ends debates rather than starts it, it kills curiosity in organizations.

And, one of the things we took away from what Jeff Bezos talked about was treating anecdotes as data. When an anecdote, that, that whisper, that customer comment that's in an email, that slight discomfort that there's an outlier, when the anecdote contradicts the data, it's often the case that the anecdote is right.

Not because the data's wrong, but maybe we're measuring the wrong thing, or maybe the environment That we were measuring has changed. So that single angry customer email, for instance, should trigger more more inquiry, more curiosity, not get explained away by an average number. 

Richie Cotton: Yeah, it's definitely true that averages don't always tell the whole story and yeah, you should pay attention to what your customers are saying.

Do you think that there's a way to make sure that your colleagues in general are using data in the right way then? Like, how do you go about making sure that people understand when should you be using data? When should you be paying attention to one thing over another? 

Jana Werner: We argue for maybe shifting the mindset just a tiny bit, and I think that's already what helps to do it differently.

It's going from data, from the idea of being data informed rather than data-driven. I think once you already have that mindset, it shifts the voice of the conversation. And things we see done really well is when you, for example, move your data analysts alongside domain experts because then you suddenly get context.

So you don't run a central team that has silos and fields requests, but you... The, the data scientists join the daily standups or they actually become embedded in a team, and then they gain context and that, that kind of stops mathematically sound models from missing the obvious that seems obvious when you're close to what's happening in the world.

So there's lots of little things I think you can do to put things in context and then use them as a leader. You're data informed but you also use your wisdom, your instinct, your background and the anecdotes that, that Phil talked about. 

Richie Cotton: Yeah, I do love this idea of making sure that your data teams are very close to your business teams, so you've got that domain expertise and your technical expertise together and people are actually having conversations that they're gonna be greater than the sum of the parts.

Okay. Are there any particular data skills that you think are broadly useful across your organization? 

Phil Le-Brun: Yeah, and I think it's... They're actually skills that are becoming increasingly important with AI. Critical thinking, for instance. Does... What's the data telling me? What does the metric... What can the metric tell me and what can't it tell me?

When should I look at averages? When should I look at outliers? How do I ask questions of the data rather than just take it at face value? And I think we see the same with large language models today. The one report said that 60% of people take the output of a large language model and accept it, which is crazy.

Being able to have a debate using the data to critically inform your point of view is really important. And, we know, we've known for years one of the biggest challenges with data is the low level of data and technology literacy from the senior leaders on down. It's the inability to use data for what it's intended.

So maybe I can tell you how many Big Macs I sold last Thursday, but can I- Am I any good at predictive and prescriptive analytics? How many Big Macs am I gonna sell next Wednesday, and what can I do about it? So giving people the confidence to use data, but also that critical thinking so they don't just blindly follow the data.

Richie Cotton: Absolutely, yeah. Going from that kind of basic descriptive statistic to actually getting the, it to be something useful as part of a decision, that's it's several extra layers of skill beyond that. Okay so there was another anti-pattern which is a little bit related to this, which is about seeking perfect decisions, which sounds like a good thing.

Why is this an anti-pattern? 

Jana Werner: Because chasing a perfect decision is a really expensive habit, and it's actually not helpful because the fear of getting it wrong and keep searching costs you more than just taking a decision and learning from the outcomes. You gather more and more data, and that actually also surfaces more options, not fewer.

And then what happens to our human brains is the more options we have there's more paralysis. I love this really simple experiment someone did in a supermarket. They offered on one table, I think it was 24 jam types, and on another table they offered customers six jam types. And at the end of the day, they found that the table with just the six jam types sold multiples compared to the one with 24.

So you think actually more choice is good, but it's worse. It sends our brains into a kind of defensive crouch, and the safest option then is to not choose at all and wait. So less choice and moving forward is really helpful. And the coolest thing we've learned here is a trap we've learned from Annie Duke.

She's the most successful female poker player in the world, $4 million prize money richer, but she also has a PhD in decision sciences, and she talks about resulting. Resulting means that when you judge a decision by its outcome you feel that if you had a good outcome, it was a good decision, and when it was a bad outcome, it was a bad decision.

We know that's not true. The world is complex. You can make a great decision and have a bad outcome. So think about this. Understand what resulting is. Don't pile on more layers. You'll just end up with slow, mediocre decisions or no decisions like the 24 jam jars. And just think about that it's okay.

You might make really good decisions and sometimes have bad outcomes, and that's fine, but learn how you make good decisions. 

Richie Cotton: Buying stuff there's so many weird behavioral science decisions about how people go about making purchases and things like that. And I do think, yeah, it's important to think about, like, how do you make sure that you are not over-egging things, i- is the goal of this.

How can you go about having a process to make sure that you're consistently making good decisions then? That seems to be the crux of it. 

Phil Le-Brun: You can't. It- we live in a world where there's more questions than answers today. But the most important thing is you actually make a decision because as soon as you make a decision, you can start to gather information on whether it's a good decision or not.

We spend so much time paralyzed trying to make the perfect decision, that no decision is the worst possible outcome. And then, all, not all decisions are gonna be good. So that willingness to be able to change your mind quickly, that's really important, too. We have a leadership principle at Amazon which is, "Leaders are right a lot."

And it doesn't mean what people normally think it means. It means that if there is some disconfirming information that says your decision was wrong- We're gonna change our mind. So at the end of the day, we're right a lot because we're willing to adjust based on the data. So instead of agonizing over a single best option, we tend to rapidly sort th- the possible outcomes into good enough to try or not worth pursuing now.

And then we use a two good option rule. So the moment we have two solid viable choices, we stop analyzing, and we ask if we could only have one of these, would we be happy? So if the answer is yes for both, let the decision owner choose one of them and just try it. They can always change their mind. They can always go down a different route after.

So it's a difference between clarity and control to us. You don't prevent poor decisions with more bureaucracy, more gates, more governance. You prevent them by not being clear about which door you're gonna walk through and who owns the call. And y- goes back to make a quick decision, but equally be prepared to change your mind quickly if there's disconfirming information you find.

Richie Cotton: I love the idea that most of the time it's a good idea to make a decision and then you can always change your mind later. But there's a whole thing Jeff Bezos had this spiel about one-way, two-way doors, but some decisions are reversible and some aren't. I guess that makes the decision a, a difference to this decision-making flow, right?

Jana Werner: We believe that most decisions actually are two-way doors, so you can go through the door and you can come back, like adding a button on the amazon.com website. There's some one-way door decisions like if if you build a big superstore somewhere physically that's a one-way decision. It would take a lot...

you could still reverse it, but it's very expensive to reverse it. And so- The idea is to get people in this, using this really simple visual to think about how many dec- how dangerous is this decision really? And if it's reversible take it with about 70% of the information you wish you had.

That feels so uncomfortable and you have to train like a sports athlete to get used to it and hold your discomfort and go for it. And when you do it a few times and there's a culture where it's okay, and this is acceptable and actually desired, where it's a failure if you keep waiting to be a perfect decision maker, then this becomes a habit and infuses massive speed in your organization, and you learn a lot.

You get a lot more information. 

Richie Cotton: Okay. Yeah, I love that. So you're focusing on going faster and it's okay, you make mistakes, you can always change course. You do the famous startup style pivot. Yeah. All, all good. Okay. All right we talked a lot about decision making. I'd now like to switch and talk about hiring.

One of your anti-patterns about hiring poorly is why is it so hard to hire good people? 

Phil Le-Brun: I don't think it is hard to hire poor- good people. Often we hire good people and get in their way, and we impose all these levels of bureaucracy on top of them. We had a colleague Adrian Cockcroft, he used to work at Netflix, and one day he was on stage and he was telling a story about what Netflix were doing, and he opened it up for Q&A, and one of the audience members said...

It was less of a question, it was more an accusation, said, "It's all right for you. You hire the best people." And Adrian turned around and said, "No, we hire your people, we just get out of their way." And I think a lot of this comes down to you're hiring you're hiring people because they're brilliant.

They know something. They're intrinsically motivated. They want to do a good job. That's what most human beings are like. But then we make it really hard for them to do a good job. 

Richie Cotton: Okay, so it sounds like there's a gap, then, between I'm hiring and then I'm making them successful. What's that gap?

Jana Werner: The gap is, first of all, how you hire and then how you make people successful. We often treat hiring as an inconvenience for our leaders. They have to write job descriptions. They do it side of desk. They do a bullet list of skills. We scr- scan our CVs, usually now with AI, and if there's a keyword missing, that's not the perfect person.

We also try to hire unicorns that have 30 years of AI experience. Nobody has. And then the interview becomes an hour of lying to each other, basically, as you're trying to sound really cool, and then you go, "Okay." And we don't even train people to interview well. We ask these questions that will always, that people will be prepared for "What is the one thing that you failed and learned from?"

Of course, I have a cool story because everyone asks this question. It doesn't really tell me if people have learning agility or if they're humble, and all these things. And and so that's really difficult. So a better hiring flow that we have seen is companies like Draw Inc., like Pivotal, where you actually start meeting the whole team.

At Amazon, we do loops where you have people involved that would work with you and people who would be your internal customers, where you maybe do something together and this is a polite and kind way of finding out both ways whether this works well. Or where you learn how to do interviews that you don't...

Even if you don't realize, try to hire someone who's like you, where even, for example, an AI tells you, "Look, you've just exposed bias here. You probably didn't notice, but you were bi-" And you go, "Oh my God, yes, I did." So these things can all really help. And then the next bit is really genuine onboarding.

Not like a, "You're here. There's a desk. We hired so late the, the project is already four months in. Thank God you're here. Go and fix it." That's the, that is what happened to me a lot, and that's just not the way to do it. Amazon has a much kinder way of doing that, which I really loved. 

Richie Cotton: Okay. Yeah, so many things can go wrong there.

It's very common that you're hiring because you are understaffed, and therefore you are busy as a person, and somehow you're playing catch-up before you've even started writing the job description. So yeah, I like the idea of thinking things through. Seems like there are a lot of small fixes, like, throughout the process that can change.

And I guess the, the second part is after hiring. It- it's the, the onboarding to make sure that people can be more successful. What does good onboarding look like? 

Phil Le-Brun: I remember joining Amazon. I said, "Yeah, what's my first week gonna look like?" And my boss said your first three months you're just gonna learn about Amazon."

And it's just are you joking? I thought I was gonna be thrown into the deep end like Jana, but we, the priority in the first three months is to go through a program called Embark. But if we look at how companies often treat onboarding, it's an, it, it's another inconvenience. You have to go through compliance training and a bunch of basic training.

Companies that do this well- Build a individual plan for the individual who's joining. They map out the training, who you need to meet. They protect that time so the onboarding doesn't become secondary. It is actually the primary role of that new starter over the first few months. They have a leader as a learning buddy.

They also accept, sometimes it's not gonna work out, and using this as an opportunity for that individual really to learn, what's it gonna be like working for this company? If you only do one thing differently, the one thing we learnt was add a simple three-month check-in with the hiring manager.

One of one of the exec recruiters we've worked with in the past calls it a demonstration of care. It's small, but it sets people's mind on whether to stay, and that flows through to things like retention and revenue. It's a demonstration of genuine care. I want to make sure you're going to succeed.

So onboarding isn't about getting people to productivity as quickly as possible. It's making sure they're set up for success for the long term. 

Richie Cotton: Yeah. I do have an idea of focusing on the long term, and just the idea of having that feedback loop where you speak to the hiring manager, and at least that's gonna give some information to the hiring manager to be like, "Oh, what happened to this person that that I said yes to?

Was that a good idea?" Okay. There, there was a related anti-pattern to this about the going on from onboarding in general is it was about downplaying skill development or downplaying people development. Talk me through, like, why don't organizations invest enough in their employees?

Jana Werner: It's often because they treat learning as a luxury, not the real work, even though actually nowadays learning is the real work. And they do it, what I've learned in the English language, as side of desk, if the budget allows and the time allows, and it's the first thing that gets cut when the budgets get tight.

It's not intrinsic to many organizations. Even so since the '90s we have the learning organization and all those from Peter Senge, but it's just not coming through yet. A- and there's an old tin man mindset under that, If you don't have the talent, instead of developing the skills, you think you can just buy it.

And we've heard this, this thing that sits underneath this fear many times, that if you train people, we will just train and invest in them, and then they'll leave. But actually you need to think about it the other way. You have to worry about the other thing. If you don't train them and they'll stay, that's even worse because their skills will atrophy, and you will have a workforce that isn't ready to continuously learn.

The second thing is how training gets done. We get really frustrated with this. We speak to HR departments now that say, "How do we upskill everyone for AI?" And then they invite some kind of training program, and they roll it out onto people, and that's totally separate from their daily work. The people that do this well, I spoke to a VP of, I think, data at Airbus, a quite a kind of tin man organization in parts.

But he actually did it totally different. He s- he gave his teams problems to solve that got gradually harder to go with the technology development. He always said, "You've got one to two weeks to upskill to get ready for what I'm giving you, and I will also give you space during to, to learn, and it's okay to fail and learn."

So he brought both things together. I don't know why we do this more often. We do the training of flavor of the month. We send people away. I think, Phil, you did total quality management. I did safe training. Then you come back to your company, you go, "I can't apply any of this because how the organization is set up doesn't even let me do."

So worst case, people then even come back and get more frustrated from their training than actually happy that they've learned something. So I could rant about this forever, so I'll stop here. But there's a million anti-patterns we could pick apart, yeah. 

Richie Cotton: Oh doing learning and development programs well, it's a real art.

I love the idea, yeah, you give people advance notice about, "Hey, this is the training you're going to do," and you make sure it's related to their job, and they've got time to actually learn things. Yeah, there's a big difference between that and, "Oh, here's some compliance training. You must do this by midnight tomorrow," sort of thing.

Yeah. I like that. Okay. Beyond this, do you have any other tips for making sure, like, all your employees are keeping their skills up to date? Particularly in the technology area where things are just moving quickly at the moment. 

Phil Le-Brun: We take inspiration from the octopus.

The octopus goes on a one-week course to learn to open Of course it doesn't. It's the, the... If you watch an octopus open a jar, it figures it out through cha- trial and error. It's, and it then carries that knowledge forward. One of the things that we really like, and we discovered this very, fairly recently with a few companies, is it's much of AI is about habit formation.

It's about rather than trying to bolt it onto your job, it's about how do you weave it into your daily work life? Jana and I both use AI extensively now, but it's natural. It's a habit. It's not because we've been on a training course, it's because we've developed these habits over time.

We've seen some really simple low-cost programs companies have put together to, to develop these habits in individuals. The other thing is learning happens in the workplace. About 70 to 90% of real learning happens in the workplace. So how do you turn retrospectives and one-on-ones into development conversations or debates instead of bureaucratic chores we all have to go through?

How do we articulate the types of skills we need in the future so individuals are clear about what does success look like? This isn't a... The, the technology skills are almost the simple ones to learn. What we value more is those hard soft skills, those behavioral skills, the ability to work in a team, to form a team, to give each other feedback.

They're absolutely essential in organizations today. Even coaching. So leaders need to become less less about mentoring. "Yeah, here's what I did when I did your job 20 years ago," 'cause that job doesn't exist anymore. It's more about, well- I don't understand as much as you do, but what have you thought about?

What options have you experimented with? What have you tried? Who have you talked to? So it feels a lot softer, but teaching people to coach is incredibly important. And given we live in a complex, not a complicated world, this isn't a case of I've got a question, here's the answer. It's what's the second, third, fourth, fifth order consequences of that decision.

So it's the difference between teaching someone to think systematically versus systemically. So thinking of things as a system where you can't predict all of the moves, and helping people get comfortable with that. 

Richie Cotton: Okay. Yeah. So I love the idea that rather than in your one-to-one meeting, you're going through every st- like status updates you actually you really want to spend your time as a manager making sure that your employees are thinking.

And y- you mentioned the difference between a, a complicated world and a complex world. Do you just wanna expand on that? 

Phil Le-Brun: Complicated is like a bike. You take the wheel off, you read the manual, you figure out how to put the new wheel on. Our organizations aren't l- like that. You can't unplug waterfall methodology or the marketing department and plug in agile or an outsourced marketing department 'cause us humans aren't too good with that.

So com- complex organizations, complex systems, which is where the octopus metaphor fits in you can't just analyze a human body or an octopus body and figure out how it all works. The history of medicine is you try something, it has a positive effect, it have, has an adverse effect. You try something else.

You learn. So you can't possibly predict all of the unintended consequences, and everything you do has a consequence you can't predict. For instance, when we looked at what it would take to embed full agentic worklo- workflows into an organization, there's over 250 changes typically required to your governance, to how you hire, to who runs the workflows.

So you have to just learn and iterate your way through rather than trying to come up with the perfect plan. 

Richie Cotton: Yeah. I think a lot of people have the sense that the world is somehow quite uncertain at the moment, and I think yeah just being able to be comfortable that where you go, okay yeah you're not gonna be able to predict the outcome of everything you do.

You're not gonna be able to understand absolutely everything, but, you make your way through it. 

Jana Werner: Yeah. I think this is a really important point, Richie. We... You no longer can say I have a vision where I want to change to," and take people with me on that journey, where the job of leader was to take people with you.

It's now a nobody knows. I don't have the answer, and all you can do is invite people to come on the journey with you and create an organization that can really quickly figure out and learn what works and what doesn't, and keep learning. That's, I think, the biggest difference we try and help leaders understand right now.

Richie Cotton: Okay. Yeah. I love that. So- We talked about quite a few anti-patterns Suppose you've identified these problems within your organization, how do you go about changing your business to get rid of those anti-patterns? 

Phil Le-Brun: The tendency is for the leaders to come together and try and figure out all the things that need to change which is normally everything.

So then they launch a big transformation plan, which we know doesn't work. Yet, there's simple things we can do. Just imagine you could light 1,000 fires in your organization, and this is a quote from a chief transformation officer of a mining company, which isn't... that industry is not renowned for transformational innovation.

His point was if you can have 1,000 people come into your organization come into work every day thinking, "I can make a difference. I know what the company's purpose is. I buy into it. I have the agency to make a difference. I can express my curiosity without fear of being shut down or penalized." If you create that sort of environment, it can be as simple as you as a team leader sitting down with the, with your team, flicking through the book, and watching your team's reaction when they read something, where their shoulders grow heavy, they roll their eyes, they start to laugh.

They know where these issues are. They've just never been given the agency, the permission to do anything about them. Tap into that. You set... Light 1,000 fires. Let people experiment with their own role with their own job in their own team, and try things. Just as we do with experiments with software, for instance, do the same with your own culture.

It's amazing because it also has this motivational effect where it gets people more passionate about their job, more bought into the organization, because we all want to feel like we're making a difference, not just doing a job description. 

Richie Cotton: Okay. I love that. It sounds like a lot of that then is just about as a manager make sure your employees feel empowered and that they wanna show up to, to work a- and do their best work then.

Phil Le-Brun: Yes. Although I think often we s- we say things like feel empowered, and that's probably an accurate description. Empowerment to us is always, "Hey, I'm gonna give you permission as long as you don't make a mistake." Or, "And when you make a mistake, why didn't you come and talk to me?" That's not em- So we often talk...

we feel that often managers dish out empowerment on the assumption they can take it back again. So when we talk about ownership, we think about responsibility and authorship, this, this idea that I'm responsible but I can also decide how I'm going to approach a, a particular problem. And all of this has to start obviously with a clear vision of what the organization stands stands for, and often that's missing in the majority of organizations we talk to.

Richie Cotton: Okay. Yeah. So it sounds like there's a lot of small steps then towards change. Jana, do you have any more tips on, like, how do you go about making these cultural changes to your organization? 

Jana Werner: Yeah. One thing I'd recommend is as a leader we are designed to add something, to make something, to create something.

I look at it the other way. What can you take away? What is one thing you can subtract from your organization- organization, a process, you can get rid of it. What is something that we learned from the CEO of of the South African Stock Exchange? Use the hell yes test. If all your people sit around table and don't say hell yes, maybe that's something you shouldn't be doing if there's not excitement.

And don't just set priorities as a CEO. Think about what you take away, how you create focus or something and- our point is, what we want to say is, this doesn't have to be hard. Everybody can see anti-patterns in their organization when people start rolling their eyes, or their shoulders go heavy, or they giggle, or they go, "Ugh."

Then you know there is an anti-pattern there. Just follow that compass and light what we learned light a thousand fires. And someone from a South African mining company said, mining's quite a traditional industry, but he said, "I try to light a thousand fires." It means that everybody can help change.

Everybody can pick an anti-pattern, try something, and you will have always learned something. Maybe you haven't always changed something, but you uncover something. And this becomes no longer the heavy burden of the leader. It becomes the e- the enabling thing that everyone can help with. 

Richie Cotton: Light a thousand fires, but not while you're down the mine.

That's very dangerous. Okay. Yeah. So I like this idea that if your employees are saying, "Hell yeah," then it's probably a good idea. If they're rolling their eyes and shrugging their shoulders, it's probably a, a, a bad thing. All right. Before we wrap up, I always want more people to learn from, so who's work are you most excited about right now?

Jana Werner: I would like to pick two. One is Linda Hill, a Harvard professor. She's about to get a lifetime award on innovation and leadership. And sh- in this current time, she has a beautiful metaphor she's developing for web- with leaders and everyone in organizations going from wayfinders, where you know where you're going, to...

Sorry, from pathfinders to wayfinders. Like these, these discoverers that, that went around the world and discovered new land. I think that's really interesting, and she talks about people that are bridgers, that can bring together siloed organizations, make things happen together. We need this more than ever.

So it's really exciting what she's talking about. You should definitely have her on your podcast. She's mind-blowing. And the other one is, we just met him, is Michalis Annini. He, he wrote Human- co-authored Humanocracy, and he's now working with Bayer, and he's completely changing how people work together, the structures.

He creates more than an octopus organization in real life, in a big big company, so really fascinating. 

Phil Le-Brun: I'll add one more. One of our colleagues at work Mark Schwartz, who's a former US government CIO. He's written books like Seat at the Table, a- and talks about things which sound like management heresy, like decentralize by default.

And people are saying, "No, we should centralize." And his whole point is you give agility to the edges of your business. But he talks about how in highly regulated industries, for instance, you can still innovate. You can still do amazing things. You don't need to allow compliance overhead to stand in the way.

It often becomes an excuse. 

Richie Cotton: I love it. Yeah. Lots of exciting ideas there so some great work for me to follow up on. Wonderful. All right. Thank you for your time, Phil. Thank you for your time, Jana. It's great chatting with you both. 

Jana Werner: Thank you so much. 

Phil Le-Brun: Thanks, Richie.