Julian Birkinshaw from London Business School: Bureaucracy, Meritocracy and Adhocracy

I am not going to get into the practical details of budgeting. I don’t know enough about that to do a good job for you on that. But what I can do, what I will be doing, is offering some thoughts, reflections on the big challenges facing small and large companies, the ways that they’re thinking about organising in response to those challenges, the implication for those organising models for individuals at all levels in their organisations.

So in some ways it’s a way of getting you to the world of beyond budgeting but from a slightly different starting place. I’m going to have two points during my talks, as I was asked to do, where I’m going to get you talking just for a couple of minutes at your tables about a question.

The future of management is a pretty broad and ambitious topic to cover in half an hour. I’m going to give you a couple of different ways of looking at this. This is an increasingly common framing is to say that the world is just moving faster and not just moving faster but the speed of movement is quicker. You’ve all heard of Moore's law, this idea that the number of computer chips on a wafer is going to double every 18 months or so. If you extrapolate that out, you can see any number of typically technological things which are increasing at, shall we say, exponential rates.

There’s even a kind of a theory behind this that a chap called Ray Kurzweil has come up with and this is the Law of Accelerating Returns. You don’t have to believe this because I don’t believe it, but basically it says that within the next 20 years the rate of increase of artificial intelligence, computer based intelligence, will be such that within that 20 year period we’ll have reached the singularity, which is where human intelligence and artificial intelligence are kind of as one. Artificial intelligence will be as powerful as human intelligence.

He’s got lots of arguments around that. He’s got lots of theories for how that is possible and what that would look like, and I’m not going to get into those, but I think it is worth both acknowledging the potential there in terms of what artificial intelligence offers us whilst also acknowledging the limitations of that point of view.

And the limitations of that point of view are actually best expressed with this different hypothesis, which I’m going to call the creative destruction hypothesis, which says that whenever lots of stuff changes, the opportunities that creates for us as organisations actually, the substantial nature of that, changes as well in ways that perhaps you wouldn’t predict.

In order to create something new we have to kill something off in its place. In order for technological progress to advance what we have to do is we have to understand that there will be winners and losers in that process. So the companies that win in the future won’t necessarily just be the ones who have grabbed onto technology and turned that into an asset.

Let me explain what I mean by that. Here is a very, very simple framing of the transition that we as a society went through 50-odd years ago; the transition from the industrial age to the information age. That’s not a contentious framing of the problem, but it is worth speculating just for a minute as to what that transition in the 1960s-70s looked like, because that was the period when so-called industrial age companies gave way to so-called information age companies.

The way that I frame that transition is I say that the classical laws of economics said that capital and labour were the scarce resources, but gradually as information became more available and more useful, the possibility of using information and knowledge actually allowed different types of companies to succeed.

So the companies that succeeded in the industrial era – think of the big traditional Volkswagens and General Motors’ and Fords – they succeeded by squeezing capital and labour to make the most out of them. But they actually ended up giving way to other companies – Toyota within the car industry of course was a leading adherent of this, but increasingly other companies in other industries as well – by essentially figuring out that no longer were capital and labour the scarce resources.

Actually there was plenty of capital around, there was plenty of people, what matters was what you did with this stuff. And the companies that were thoughtful about how they put capital into longer term uses and the people who were thoughtful about how to get the brains of their employees to use as well as just their hands and their feet were the companies that won.

When we look at today’s business world of course the companies that have succeeded are all knowledge or information-based companies. That’s really just a backdrop for my first little teaser question, which is the following:

Is it possible that the information age in which we now live will run out of steam? Is it possible that in fact we won’t live in a world of accelerating change forever? That the companies that are really good at harnessing information will actually at some point stop being the leaders? That something else will, shall we say, take its place? Is it possible that we can imagine looking back from an age that lies beyond the information age – we’ll call it for the moment the post-information age – where the companies that succeed are not the ones which are just obsessive about harnessing and managing and manipulating information?

I realise this is a little bit challenging or unorthodox. We live in this era of big data, this era of business analytics, where a lot of companies make their livings by selling information and the capacity to manage information, but I’m going to suggest that whether it’s now or in ten or 20 years time, we actually have to look beyond that. There’s a clue in the term “information age” and age is a period in term, which starts and finishes.

I’m not saying for a second that we won’t need information in the future. We will always need information. There will always be value in information just as there was always value in labour and capital in the information age. The question should be instead, what will be the sources of competitive advantage for companies in the future? In other words, what would a world look like where information was no longer a scarce resource, where information was ubiquitous? I don’t think we need to think very hard about that. The search costs today are low. Information is everywhere.

What does a world with actually too much information look like where no longer do we have to worry about getting hold of this stuff? What become the scarce resources of the future? Essentially strategy and management is about making the most of scarce resources. The scarce resources of that era have given way to the scarce resources of the current era – information, knowledge. What will the scarcest resources of the future be?

Here’s my first little teaser question: what would be the warning signs that the information age was coming to an end? What would a business world with too much information look like? What problems does that create? How might we resolve those problems?

The second [question’s] a bit more puzzling: what would a business world with too much emphasis on knowledge look like? What problems does that create? If we can answer those questions cleverly it actually helps us to figure out what the winning companies of the future will look like, the ones that have gone beyond or transcended the information era.

Three minutes on this question – go!

I want to get two or three observations. Table at the back, what do you guys think?

Delegate 1: Civil liberties.

JB: Yep, so there’s a massive civil liberty question out here in terms of secrecy and right to privacy and all that sort of stuff.

Delegate 2: We were saying that in a sense we were already there. Information, data, people just give it away. Kids don’t care about it. Kids just put all this stuff on Facebook. So in a sense it’s not a scarce resource anymore

JB: Correct. So if we’ve got these layers with data at the bottom, information above, knowledge above that, wisdom above that, the basic data is more or less free. It’s very, very rare that any company can genuinely put its arms around a body of data. There’s enormous volumes of it. We call it big data for a reason, because there’s literally billions of individual items. That does therefore put a premium on the capacity to do something with it.

We can start with agreeing, I think, that basic data, basic information is ubiquitous, is everywhere, is more or less free. So what problems does that create at a corporate level?

Delegate 3: Well it’s not as scarce anymore.

JB: Indeed. So given that that’s true, how does any company then make money, because in the past it was possible to actually make money by sitting on a proprietary body of information?

Delegate 4: We were just starting to think maybe the thing that’s so different is to have relationships.

JB: Right and there’s something about a relationship which cannot be just bought and sold in the way that a body of data [can be]. There’s something about that relationship, you can get into it, but it’s to do with a trust base or an emotional connection as well as a commercial thing. Other thoughts on this?

Delegate 5: My bet is that the future is relevance and meaning, because the cracks in data are already showing. If you read all the McKinsey series on digitising your service organisation, what you realise is that there’s very little evidence and the failures are coming through fast. So they’ll be jumping onto another new thing.

JB: So one wave of the future, if you like, is creating meaning on top of this ubiquity of information. Another point is speed of hopping from one thing to the next to the extent that anything I create can be rapidly copied by others. My ability to keep moving is also potentially a source of advantage.

Others, please, on either question? What are the other risks of too much information?

Delegate 6: To take this literally, a warning sign could be when we stop discussing about data security and privacy because you can’t hide it anyway. Or when we stop wondering about what is the next step. Like what is the next system after SAPPC to access my data? If you don’t worry about this anymore.

Because you don’t need to train people to use a tool – not only data, the information you need is there. If we stop wondering or worrying about that, this could be a warning sign.

JB: What would be the internal manifestation of this sort of mountain of data that’s sitting on our computers and our collective systems? How does this problem manifest itself in the meetings that we have? Or in the internal conversations?

Delegate 7: I thought an index of how many patent lawyers country-by-country, industry-by-industry would be a very good signal of inefficient innovation.

JB: Interesting. So what are patent lawyers? They’re losing sight of the forest for the trees. I imagine these guys make a fortune, but they’re somehow missing the bigger picture.

What does too much emphasis on knowledge look like? Is it possible to have too much emphasis on knowledge within an organisation? We’re so schooled in this notion that we should be in a knowledge based world.

Delegate 8: Having savoir-faire.

JB: That’s right. So there’s different types of knowledge aren’t there. Having savoir-faire, having savvy, is actually a very different sort of knowledge than having the academic training, the logical, rational scientific training.

That’s one of the directions I want to push us in, which is that we can’t have too much knowledge. It’s almost tautological, it’s almost nonsense as humans to say that we could have too much knowledge. But we can definitively have too much emphasis on one type of knowledge at the expense of the other.

Here’s my simplistic way of answering the question. The answer is 42. I think most of you know what I’m talking about here. Very famous British novelist Douglas Adams wrote a whole series of books, The Hitchhiker’s Guide to the Galaxy being the most famous. And somewhere in that book they asked what’s the answer to the ultimate question of life, the universe and everything?

They posed that to the computer and they came back seven and half million years later – obviously it was passed down through generations that the computer was only given the answer – and the answer the computer came up with was 42. For me this symbolises neatly in one number the problem of big data, the problem that thinking somehow information and knowledge is the answer.

What’s the problem? First of all it’s overly precise. It’s exactly wrong rather than roughly right, if you see what I mean. It took seven and a half million years. It didn’t stop processing until it got to exactly the right answer, except that it’s a completely useless answer. The second reason why it’s a problem is it lacks context. What does 42 mean? It means very, very different things in different contexts.

Just two weeks ago I was in San Francisco talking to some people at Oracle and we had the guy who is in charge of Oracle’s acquisitions in the meeting. So he’s the senior vice president. His job is to essentially figure out whether Oracle should buy JD Edwards or PeopleSoft or whatever the company is.

He comes up with a number – this is the amount of money Oracle should pay and not a penny more. Then you’ve got a chap called Larry Ellison who’s the big boss of Oracle. He still owns a big chunk of the company. Larry Ellison then basically decides whether to take this guy’s advice or completely ignore it. And Larry Ellison does that on the basis of context. He decides for whatever reason that he doesn’t care what the number is, he is going to do it anyway because it’s a competitive situation, because he’s trying to take out some other potential competitor.

So we need context before any sort of number means anything and of course it’s also an answer in search of a question. Douglas Adams himself, if you re-read that little paragraph in the book, the computer says the answer’s 42 – that is the answer – the trouble is you never told me what the question was in the first place. You hadn’t ever figured out what exactly you were looking for.

So problems with big data are those three things (overly precise, lacking in context, an answer in search of a question). If we then take it back to my original framing, what is the scarce resource in a world of ubiquitous information? It turns out that actually it goes right back to the mindfulness section. Too much information creates a deficit of attention.

In a world where information is ubiquitous, we have so many stimuli crossing our desks. The scarce resource is actually not our own time. It is actually our capacity to focus on the right stuff at the right time. That has implications for time management, for how we manage our days, but it also has massive implications for how companies work.

We could easily have put the term analysis paralysis in there. In a world where you’ve got too much information, it is so easy just to wallow in that information, to get stuck in it. My students when I’m teaching them at LBS, when we run a case study discussion there’s always someone who says, “Well we need more information. We can’t answer the question. We can’t make a decision on the basis of this ten page case study.” And my answer is “you never have enough information.” There’s always going to be something else you want, but sooner or later you’ve got to actually decide.

So what’s the scarce resource? Decisive action. The scarce resource is our capacity to make decisions in a decisive way at the point where we’ve got enough information rather than when we’ve got all the information. The point about knowledge is exactly to the gentleman here, which you made the point about savoir-faire rather than academic knowledge. If we emphasise traditional academic knowledge – we’re obsessed with the idea that knowledge, somehow in a business context, can be boiled down to numbers and charts and graphs. Well, not so, right?

There’s all sorts of information which we can’t quantify. I actually think that we run the risk of making our decision-making processes sterile, lacking in meaning and that therefore puts a premium on what I’m going to call emotional conviction. In other words, a greater understanding in the underlying meaning of something, a recognition that our brain has two parts to it and that if we obsess on the left brain thinking, the rational part of it, we miss the right brain thinking, the creative, intuitive bit.

You’re going to say, these things have always been important, decisive action and emotional conviction have always been important. And they have. The simple point I’m making with this chart is, as we obsess about the big data movement, as we obsess about business and analytics and we allow ourselves to be seduced into thinking that somehow quantifying things, putting artificial intelligence behind things is the solution, we have to remind ourselves even more than usual that actually an awful lot of stuff happens through relationships. Which of course have an emotional component to them and an awful lot of it’s about knowing when to act and when to gather data.

Put it all together and this is my rather simplistic framing. I’m calling it the agile age. This is not the right word and I don’t know if there is such an age. What we can absolutely do without fear of contradiction is to put these in layers on top of each other. I think to put them in a serial form like this is overstating the point. I’m overstating to make a point.

So whether we see this as a new age or whether we see action and conviction as the complements to information and knowledge, I really don’t mind.

That’s the end of part one and all I’m doing there is playing with the different changing sources of advantage in the outside world. What I want to do in part two is to link it back to management, to link it back to what you and I do in our organisations and how we make sense of those choices. You’ll see the very explicit link to beyond budgeting in about three or four minutes time.

How do we respond to this complex world? I’ve got an image of IBM there. Why? Because IBM is a phenomenally complex organisation. I’ve known IBMers who claim that there’s like three or four different dimensions of the matrix that they’re working with at any given time. And of course IBM is not wrong in deliberately choosing big clients to work on big complex problems with, but the net result is that there’s a great deal of bureaucracy in IBM.

A great friend of mine, a former student at London Business School, just came to see me last week and he’s doing very well with IBM. He’s in the Apple-IBM joint venture, which is the kind of sexy bit of the company, and even he is saying, “look I just can’t handle it. I can’t handle the amount of checks and balances in the system.”

I’m not saying this is wrong. I’m saying it is one solution to fight complexity with complexity. If we live in a complex world, we can create a complex organisation, but we know where that leads, right? This is a picture of Max Weber, the inventor of the term bureaucracy. The point I want to make with this slide is that, Max Weber, the founder of the bureaucracy movement, wasn’t an idiot. He actually was a very smart guy and for its time bureaucracy was a clever invention.

Those are the three principles of bureaucracy (formal rules and procedures create efficiency; people are given roles according to expertise; favouritism is eliminated) and they’re not bad principles. Remember, he was fighting against the orthodoxy of the time, which was that actually companies and political organisations were often held at the whim of individual charismatic leaders who did things according to their own preferences rather than by rules.

So [Weber] was a good German, followed rules – everything is done through rules and procedures. That creates efficiency. Not wrong, but in today’s world I think we can all agree it does create all sorts of problems. Bureaucracies are risk averse, they move slowly, they take power away from those on the front line, they become incredibly internally focused. People manage internal processes as if they mattered and they completely forget about the value that they’re creating for their end customers.

So, we’ve got a choice. We can fight complexity with complexity or we can fight it with simplicity. And now the link back to the world of beyond budgeting should be clear, because for me one of the things that beyond budgeting is all about is saying surely there must be simpler ways of getting things done in order to cope with a complex world. Rather than allowing people to create big business plans, let us allow people to figure out for themselves what to do on the frontline, to make their own choices and at some point to link that back to some sort of broader agenda.

So you’ve all come across not just the beyond budgeting movement, but the whole scrum/agile movement. Do we all know what scrum and agile are? This comes from the world of computer programming. It started out with a bunch of clever computer programmer types saying the classic waterfall model for designing computer systems actually just doesn’t work. I started my career working as a little peon in one of those waterfall systems, as a systems analyst. It’s a very slow and ungainly system. Nowadays most computer programmers have moved to some version of agile programming. Scrum is one methodology that fits within that suite.

So we’ve got all of these different simpler structures. You’re quite familiar with what these are, but I think we still need to find the right complementary organising principles that go around that and that’s what I’m going to do in the second part of the talk.

So what is it that lies beyond bureaucracy? What is the management model of the future? In other words, if we could acknowledge that bureaucracy in its classical form is a bad thing – not always bad; there’s always going to be times when a bit of bureaucracy is good – we need to have some words around what the alternative looks like. And I want to get into it through a slightly tangential route.

I want you to go back to the last time you were at some sort of cocktail reception with a bunch of people that you didn’t know very well. And you’re introduced to somebody and after a couple of minutes he or she says, “So what do you do for a living?” How do you answer that question? I’ll give you a range of answers, you can tell me which one you like best. You can say, “look, here I am,” and you thrust your business card into their hand and it says, “I am John Smith. I’m the marketing manager for XYZ Company.”

This immediately tells them something about you and of course we’re all drawn to the job title. Are you an executive vice president? Are you an individual contributor? There’s all sorts of euphemisms out there for I’m very senior, I’m very junior.

I was at an event a couple of years back [with a] very senior British industrialist – he used to head a large oil and gas company – and I was introduced to him in my capacity at London Business School and I said, “I’m Julian Birkinshaw. I’m a deputy dean at London Business School.” And do you know what he asked me? He said, “How many deputy deans are there?” In other words, is this actually a small part of the management team of London Business School, or is this like calling yourself a vice president in a bank where there’s about 500 of them? In other words, position matters, status matters. He was implicitly saying, “is it worth having a conversation with you or not?”

So your job title, the formal position, tells us something about you, so that’s one way of introducing yourself. Another way is to, at least conceptually, put your CV into that person’s hand. In other words, you can talk about your knowledge. You can talk about your expertise. You can talk about the things that you’ve done over the years that give you capabilities, knowledge, expertise, wisdom in doing your work. It’s about what you know and obviously there’s ways of getting that across quite gently and quite nicely.

Or you can focus on what you actually do for a living. In other words, how you spend your time on a day-to-day basis. There was a very famous management theorist, Henry Mintzberg. 40 years ago he blew apart the thinking about management by actually following chief executives around their offices and into the building sites and out to see customers. He figured out that what people talked about and what they actually did on a day-to-day basis were two very different things. So you can actually describe yourself through what you do on a day-to-day basis rather than your position or what you know.

These are three very different lenses through which you can describe yourself and they map quite nicely onto three distinct archetypes, pure models, of how organisations work. This is the big idea and what I want to do is map those three lenses onto three organisational models.

Bureaucracy. If we sharpen the definition of bureaucracy it basically says that I have legitimate authority over you. My formal position is privileged. If you disagree with me and I’m your boss, I can basically just ignore you. I can pull rank, I can tell you what to do.

What is a meritocracy? Most of us, without consciously thinking about what the word means, would probably say we work in a meritocracy. It sounds like a good thing; it’s to do with knowledge. What it basically says is, what I know is more important than where I sit in the system. In other words, my knowledge is privileged.

If I work as an academic, I’m the chairman of a department at London Business School, if I call a meeting, people choose whether to show up. If I suggest we hire somebody, people will challenge me. Unless I win the argument I don’t get my way. Meritocracy says, it doesn’t matter where you sit in the organisation, it’s what I bring to the table in terms of knowledge and expertise that’s counts.

You can see where I’m going here. There’s a third model and we’re going to call it adhocracy. Go back to the cocktail reception. You can describe yourself in terms of your formal position in the org chart, in terms of what you know, or you can describe yourself in terms of what you do. In an adhocracy, action is privileged over knowledge or position. The term adhocracy is not mine. It’s been around for a few years. We’re reinventing here and giving it a sharper definition.

What we’re saying is that imagine that you’ve got a sales person who’s on the phone to an unhappy customer. That sales person has three choices about how to deal with the unhappy customer. One choice is to say, “I don’t know what to do here. I’m going to get you to talk to my boss.” That’s going about the bureaucratic route – you’re following the rules and your escalating to your boss.

In a meritocracy that person might want to get to the bottom of the problem – “help me understand what it is that’s gone wrong here? I want to figure it out.” And then the adhocracy, the sales person says, “my job is to make sure that as quickly as possible we resolve this issue and we move on.” Those are very different ways of dealing with an issue.

These are actually very, very different worldviews if you think about it. The bureaucratic model says we’re coordinating things through standardised rules and procedures. We’re making decisions on a top-down basis. We’re actually motivating people through the extrinsic rewards by actually paying them for results. Very different logic than a logic built on a meritocracy-based approach versus one that’s built on adhocracy.

What does an adhocracy look like in practice? If you’ve ever worked on the trading floor in a bank or perhaps you’ve worked on a deal where we’ve got to work all night to pull the materials together before doing a deal for an acquisition, that’s adhocracy at work. If you’ve ever worked in, for example, a Skunk Works operation, one of those research and development units off to the side of a company. If you’ve ever worked in an operating theatre or an emergency room.

In all these cases, we’re not saying that knowledge doesn’t matter or that position doesn’t matter. We’re saying that, on the margin, if we have to choose we are better off doing something than either thinking and debating it or deferring to our boss.

Let me just spend two minutes going a bit deeper on the adhocracy and then I’m going to kick it back to you and just say, ok, which of these models do you recognise as things that you’re using and which ones would you like to use more of?

In an adhocracy we coordinate not around processes or procedures, not around an academic discussion, a mutual adjustment. We coordinate around opportunities. Have you ever heard of Valve? It’s a West Coast American gaming company created by a guy who came out of Microsoft, Gabe Newell. It’s got 300 people [and] it’s got no managers.

I’m not advocating the no manager route. I think there’s dangers in having too few managers. But what I do like about his model, this is the employees’ manual. When you join Valve he says, “You were hired to be constantly looking around for the most valuable work you could be doing.” You have no job description. You’re job essentially is to find something useful to do.

Now, with 300 people we can do that, because everybody just about knows everybody. Gaming is one of those industries where actually you can create a very small team and create a prototype game literally in a couple of days. And the downside risk of getting it wrong is also pretty low. We’re not running a nuclear power plant here. What we’re doing is creating little games for kids to play on.

We have a particular situation here, but it actually illustrates the bigger point, which is that the mechanism for coordination is the opportunity that we see in the market place.

A second thought on what adhocracy looks like in practice. We all know Costa Coffee. This is a picture of their third generation vending machine. This is the one that, as you approach it, first of all you see it’s beautifully designed – it was designed by Pininfarina in Italy. Secondly, as you approach it starts making noises, the coffee shop in the background noises and it starts wafting scents of coffee and vanilla flavour over you. And then you’ve got this enormous great screen, which starts to do a little bit of facial recognition to decide which sort of coffees to offer you. It’s beautiful.

The key point is the following: for Costa, which is a company that has become famous for its coffee shops, this is a little bit of an unusual thing. Is this in some ways going to cannibalise their existing business or is it going to complement it? They didn’t know for sure, so what they did was they created one of these classic Skunk Works teams and I followed this process very closely. I got to know the guy Eric Actman who runs it.

They did the whole project from a standing start to a low-rate production model in eight months flat, nine different partners around the world – there was partners in Switzerland, France, Germany and so forth – and with a budget of less than a million dollars because they got a lot of these partners to put their development work in for free on the basis that they would get paid on the upside of the sales.

Classic Skunk Works operation, right, and the point is that every decision that was made was about moving forward. It’s classic entrepreneurial stuff really. We’re making decisions on the fly. Our job is to make sure that we get to the end point as quickly as possibly.

So you see that this is adhocracy at work. The challenge with that particular example is, even though it was the classic case of adhocracy, the rest of Costa was still being run broadly as a bureaucracy with a meritocracy sort of element. And as soon as they finished the project you can guess what happened. The project finished and then it bumped up against the practical realities of how do you roll this thing out? And the thing ground to a halt a little bit at that point, so all the speed to some degree was lost.

So here we have my master plan. If you buy the argument that the industrial age was based around bureaucratic principles, that’s an easy argument. And if you buy the argument that today most of us would say that the model of organising that we’re aspiring to is the meritocracy, there is a risk of meritocracy, which is that we’ll become obsessed with information and knowledge for its own sake.

I can think of plenty of companies which are so obsessed with how smart they are that they become a little bit inward looking. They become thoughtful about the knowledge they’ve created and they lose track of what it is that the market needs. Universities do this all the time. Universities have a licence to do that. It’s a problem when companies start mimicking universities.

So if you buy that there are limits to the meritocracy, something must therefore go beyond that. And that something I’m going to call it an adhocracy. See what I’m doing here, which hopefully is a little bit different, is I’m not just saying bureaucracy = bad therefore beyond budgeting = good. I’m trying to say that if bureaucracy has got its bad elements, we actually have to be a bit more thoughtful about the alternative to bureaucracy. And it’s not just one thing. It’s not just everything’s a free for all. It’s actually a combination of what I’m calling meritocracy and adhocracy.

So second little discussion thing. Which model or models are you currently using today? And the answer’s going to depend a lot for example on how big your organisation is. And secondly, which ones would you like to be using and what would be the path from A to B? Spend three minutes on this.

So lets hear your thoughts. Any observations? You guys at the back, what did you discuss?

Delegate 1: I think we all agreed here that we have a mixture of all of those systems in our companies. And also we mentioned that when companies grow they become more and more bureaucratic.

JB: Terrific. Two points. a) We can see bits of these in different parts of our companies. If I work in the bank, this is the compliance function, maybe this is the sales guys, maybe this is the analyst. But your second point, which is important, is that as a start-up company you almost by definition start on the right hand side here and then there’s this natural aging and growing process that takes you gradually over to the left side. Then by the time we’re an organisation of a thousand plus people, it’s very hard not to be a bureaucracy.

The natural aging process, if you don’t act very thoughtfully, is to allow the formal processes and systems to take over and to find ourselves driven by these processes and losing track of your customers.

Delegate 2: But I think that’s a rationalisation. You hear it a lot – as you get bigger, you naturally have to become more [bureaucratic].

JB: And you’re saying, and I’m going to agree with you, hang on do we have to assume that is a natural process? Or is it possible to arrest that process and stay close to our customers? Absolutely.

Delegate 3: There is something about… quite a lot of people have a natural sense that they like to know what their place is in the organisation. They prefer to work within structures. I know I’ve made recruitment mistakes when I’ve recruited someone and they hated it, because they couldn’t cope with the fluid [organisation].

JB: So we all have this natural predisposition toward a certain style and you can get it wrong in both directions. You can have this person who just craves the security and comfort and predictability and you throw them into an adhocracy and they go crazy. You can go to my former student who’s at IBM and he’s going crazy in the other way, which is he is an entrepreneur. He tried starting a bank three years ago and he’s being driven crazy by IBM.

One of the reasons this is useful is you’ve got to think about these models not just in terms of the rules and structures we’re creating to make them work, but also the types of people that live in them. One of the things we’ve got to get right is making sure that the people and the model are aligned.

Any other thoughts on that? There is a force of gravity, which pulls us unless we work very hard to the left. And yet, when you discussed this in your tables a few minutes ago, my guess is a lot of you were saying, “we would love to be a little bit more adhocratic, we’d like to be a little bit closer to the customers.” You wouldn’t be here otherwise.

The key point is the following: unless we work very hard at creating an organising model that helps us to do this, that actually privileges action, what ends up happening when push comes to shove is that we revert to type and we fall back on the classical model. So many of you will have been in those meeting where you want to get something done and your boss says, “Let’s reflect, let’s take some time, let’s gather some more data, let’s make sure that compliance have said it’s ok.”

Delegate 4: For me it’s not really the one or the other. Having a healthy bureaucracy and a healthy meritocracy is the basic tool to have adhocracy. It’s not you leave one behind. You need to have a certain level for healthy or good structure there to enable [adhocracy].

JB: I agree with you. The only caveat I will add is unless you’re very explicit that this adhocracy bid is allowed, people are going to allow themselves to get dragged back.

Delegate 5: We were saying, we’ve got some basic measures in place like return on investment for an investment on bureaucracy. If you have that in place in an adhocracy you’re going to stymie invention because the guy’s going to say, “I don’t know, I’m just giving it a go.” And the guy says, “Well give me your ROI,” and it’s just not going to work.

JB: Ultimately what gets measured gets done and if our measures are all built around this model [bureaucracy] or even this model [meritocracy] then not surprisingly this stuff loses traction and we end up gravitating back.

Delegate 6: Happy’s kind of a small business so we by default are kind of an adhocracy. Anybody in a small business will probably recognise that. But the experimentation piece is the one where we probably have the most trouble, because the pressure to go towards meritocracy and bureaucracy comes from trying to get some sort of certainty around experimentation.

The temptation is to jump on your horse and ride off in all directions. If you took a slightly more scientific approach to experimentation and said, “We hypothesise that if we do X then the result will be Y,” that’s not necessarily driving you towards meritocracy or bureaucracy, but it is how to experiment properly and actually understand the results that you’re seeing and make sense of them and therefore achieve something.

It’s that lack of results from experimentation, by not experimenting properly and having a hypothesis that’s good, that pressures people into meritocracy and bureaucracy.

JB: It’s a terrific point because as you say the notion of an experiment makes a lot of people uncomfortable, particularly those who are coming from this bureaucratic background. By defining your hypothesis, by putting it clearly within a box, by making it clear what the downside risks are – all of those thing provide a little bit more comfort for those who don’t really like the word.

I’m just going to spend five minutes building on this experimentation point. This is a story with a twist. It turns out that most of you have heard a version of this, but I’m going to give it a fresh lick of paint.

It is the story about high jumping and the graph goes from 1900-2000. This is the men’s high jump world record over that period of time and what you can see immediately there is that in that period – it’s the early-‘60s – that world high jump record was raised seven times. What’s going on there? Drugs is a good answer. I can’t prove it’s not drugs. Dick Fosbury’s a great answer. That’s my answer. It’s the wrong answer.

You look at that chart and you think, wow, it must be Fosbury, he’s the guy who raised the bar. Well, actually not true. It turns out that dramatic growth in the men’s high jump world record was two guys who you’ve never heard of, John Thomas and Valeriy Brumel, who jumped over the bar forward – it was called a straddle jump.

Fosbury was the guy who invented the jumping backwards. These guys were jumping over the bar forwards and they’re the ones who raised the world record in the early-‘60s. So what’s going on there? Partly this is one of those classic rivalries. This is the Federer-Nadal thing that pushes them to raise their game.

I went back to the archives and I plotted [Fosbury’s] best jumps in that same period and that’s the green dots. Way back in 1960, ok, here’s Dick Fosbury, he’s jumping a good 10-15 centremetres lower than world record pace. He’s not even an elite competitor at this stage. He’s quite young at this stage; he’s still only 17/18 years old and he’s jumping over the bar backwards.

His coaches are saying to him, “Dick stop doing that. It’s dangerous. You’re not breaking any world records.” And yet he persevered, he stubbornly continued with this methodology. It turns out that the big foam mats were only introduced in about 1961-62. That was part of the reason he was able to try this, because up until that point you really would’ve broken your neck if you jumped over the bar backwards.

So up until the Olympics in Mexico he was not really that well known. Then he got this piece of amazing luck. Valeriy Brumel broke his leg on a motorbike accident. He was never able to jump again. So at the Olympics in ’68 Dick Fosbury shows up as this relatively unknown guy and the world record holder isn’t there and it turns out that Fosbury wins the gold medal.

This is when the world takes notice. Here’s this guy jumping over the bar backwards, using this crazy technique. Everybody else is jumping over the bar with the straddle jump. When you plot the world records in the subsequent period you see red dots are old method, green dots are new method. We actually see for the next ten years, the two methods of jumping are actually co-existing.

Fosbury never broke the world record. So he was the guy who worked hard, built the technique, was given a hard time by all his friends for doing something a bit weird. He had the tenacity to stick with it and he paved the way for others. So why am I telling you this story to finish off a talk about adhocracy? Two reasons.

First of all this is the classic disruptive innovation story. You’ve heard this time and time again: the innovations which really make life difficult for you are the innovations which come from below. The Fosbury flop was a disruptive innovation because the guys who were doing the straddle jump didn’t take it seriously until too late.

But the real reason for showing this is actually the individual story not the industry story. Fosbury was a stubborn, tenacious, creative guy. He happened to be right, he happened to get his timing right, but not every single one of those guys do that. I see Fosbury as a classic example of an unreasonable person.

This is Elon Musk. This is the guy behind Tesla, SpaceX, his latest venture is the so-called Hyperloop, which is going to wiz you from San Francisco to Los Angeles in 35 minutes flat. It’s crazy and yet you can’t bet against him because he’s been right so many times before.

Musk and Fosbury are unreasonable, and I use the word unreasonable in a George Bernard Shaw kind of way. Shaw said, “The reasonable man adapts himself to the world. The unreasonable man insists on trying to adapt the world to himself.” Therefore if we want progress we need unreasonable men or women.

What’s that got to do with adhocracy is that in order for action taking to be the thing that’s pushing our organisations, we need to find ways of encouraging these sort of people. We’ve got to find methodologies in our organsations which acutally keep the Elon Musks and the Dick Fosburys happy. And that’s very, very difficult

I’m going to finish with this chart to make the link back to leadership in these different eras. Go back to where I started. What is the leadership challenge in an adhocracy? What we’re trying to do is we’re trying to enable experimentation. We’re trying to encourage people to try new stuff. That means, for example, becoming much more tolerant when things don’t work.

There’s also this need to make an emotional connection. Emotional conviction was one of my drivers. Emotional conviction works within the organisation as well as outside the organisation with our customers.

So I’m going to wrap up there. There was one question at the back around holacracy. Have any of you heard this term holacracy? A guy called Brian Robertson has come up with this term. It’s been tried out in Zappos, which is an Amazon subsidiary. It’s one example of a specific not just label around a new management concept, but actually a whole methodology for putting it in place. And arguably the holacracy is an attempt at codifying and operationalizing this notion of adhocracy.

For what it’s worth I think it’s an ambitious, courageous experiment, which may not work. I love the principle. I hope he’s successful, but I do realise that trying to enshrine adhocratic stuff in a set of rules and principles is very, very difficult.

Delegate 1: We have also been looking into that to some extent. Without going into the detail, I find that we find it very prescriptive. For that reason we struggle to see that this will really survive, to be honest. There’s one other story I’d like to share with you now to talk about this trying to fight bureaucracy. You’re absolutely right, most companies when they’re small, they’re agile, they’re flexible, they are adhocracies. They’re born like that.

It’s very often you find companies that deliberately say, well we don’t want to end up like that. But a great example is Netflix. They said right from the outset, in order to have success in our industry, we have to be extremely fast so we do not want any rules, bureaucracy to come in the way of us being successful. And there have been quite a few stories written about them. They really fought bureaucracy right from the start.

One example is that their expense policy is five words: “Act in Netflix’s best interest.” Use sound judgement. Ask someone if you’re in doubt. Keep it simple. Great example of someone who really tried hard to avoid it and then succeeded quite well.

JB: And they’ve also exhibited real decisive action in terms of aggressively pursuing video on demand.