John Atkinson on Complexity in Systems

Here, in the video (and transcript) below, John Atkinson in conversation with Stacey Hale at Design4Emergence, discusses Complexity in Systems.

Answering questions such as: 

Any advice on keeping work in a complex ecosystem within manageable boundaries?

True or false: “There are no best practices.”

How do you put people at ease in a world obsessed with big data?

You’re not saying to abandon strategy?

Are you talking about designing an attractor?

What do you say to a brand new consultant who wants to apply the Big Five of living systems to an organizational problem to create change?

How much does it matter that people know that you’re pulling from the principles of biological systems to design organizational change?


John Atkinson, co-curator of the Heart of the Art blog, shares tales of corporate change management from a complex systems lens — equal parts valor, strategy and happy accident. John is a designer, architect and catalyst for whole system change. He has instigated and led projects around the world in corporate and public settings that help people design approaches to will make fundamental change to their work and lives.

design4emergence: John, tell us more about your history with complex systems and how you apply your knowledge of the science to your work.

John Atkinson: I was thrown out of college and joined the army. That was quite a traditional form of leadership, the military. I ended up running the artillery’s leadership wing up in North Wales, and there I started to explore all sorts of different ways of understanding how you got stuff done.

I found that much more interesting than being a soldier, so I waved goodbye, handed the Queen my commission, and set out running someone else’s company. At that time I was using the outdoors to model business processes

Some people use the outdoors to develop leadership — in terms of hanging people off ropes or doing tasks. I was using it to create models that replicated what was going on in their business, so they could see it differently. But what we slowly found is it’s probably better to just work with reality than with expensive models.

I started to develop a client base that included players like Johnson & Johnson, Cummings, and IBM, as well as some European/UK sort of businesses. Without really knowing it, I was doing something quite different from everybody else. Later, it became clear that what I’d latched onto was a complex systems approach.

I was approached by the Leadership Centre for Local Government — that’s city government. Our clients were the chief executives, or city managers, and the directly elected mayor, the most senior politician and the most senior officers in the local government. What we were looking at with them was how to lead a place as opposed to an organization. Once you start to look at how places work, all your existing management theory about how organizations work starts to fall apart.

You can’t extract the organization, or the city as it were, from the place. So what is the leadership that the place needs, and what is the role of an elected leader? That starts to fall very firmly into the realm of living complex adaptive systems.

We ran a top program at Leeds Castle, which is where the G8 went. I hired Myron Rogers, whom you’ll see on the site, to come work with us in 2005. Myron and Margaret Wheatley (Meg), were the back office of somebody called Rosabeth Moss Kanter, one of the early management gurus. Interestingly, Rosabeth was a sociology professor and was on the boom of how to make business grow back in the ’70s. Myron and Meg did all the research and started to develop their understanding of how human systems work. They tied in with Fritjof Capra, and the three of them ran a retreat at Sundance Ranch several times a year about complex adaptive systems — human systems.

d4e: How did that project lead to working with global corporations, and what does a complex systems approach look like in that sort of environment?

JA: There is whole a body of thought on working with complex systems that goes back into biology and chaos theory, and that actually makes sense. At Johnson & Johnson, what’s really going on when you try and change the logistics system for Europe? What’s really going on is you’re looking at a web of relationships: suppliers, manufacturers, your facilities, hospitals, doctors, patients, and marketing people…You’re helping them see the whole web, rather than “Do we run a robotic warehouse or a human warehouse? Do we buy more trucks?” This fundamentally changed the way they worked with their business. Before, they had third party warehousing everywhere, because they couldn’t fit all their stock into the big warehouses they had. By looking at the system and how the human system and the technology system and the physical system interacted with each other, they ended up getting rid of all the third party stuff and having the place half empty, because the flows all started to work and we started to see where the connections were.

Today I’m working again with large corporate, global businesses. If you’re in financial services, the world prior to 2008 was entirely different. Today you need a different relationship with the world to survive. So what is that, and how will you adapt to an organization that can cope with negative interest rates but is supposed to deliver growth to shareholders?

d4e: How did this work lead to the Heart of the Art blog?

JA: I’d also been doing a lot of work in healthcare and local administration. For one of the clients, which is still the local administration piece, I researched and wrote a book called The Art of Change Making with Emma Loftus, who now does Heart of the Art with me, and John Jarvis who was at the Leadership Centre for Local Government in England. We asked, “Of all the work we’ve been doing with people over the last five years, what are the things that people do when working with systems, and how is it necessary?” Then we said to John, “Does the Leadership Centre want to make more of this?” And the answer was, “Yes, but we’ve got other things on our minds.”

So Emma and I decided to start again. That’s where the Heart of the Art came from, because the work was a follow-on from The Art of Change Making. It’s a bit of an oddity. We do it for fun.

But nobody was reading it because I wasn’t telling anyone. I thought somehow the world would happen across it, which of course it doesn’t. Then we thought we could use it as a site to host what other people are doing and point people at things that interest us. Could we create a place that was a resource for people who’ve been down some of the same roads that we’ve been exploring? To share our thoughts of what we’re learning as we go along, and at times to be very grumpy…

d4e: What’s the most frustrating part of working with complex systems in large organizations?

JA: There’s a piece on the blog that we got from Donella Meadows where she talks about irreducible uncertainty. All through your management career, if you’re a senior manager in a large organization, you’ve had control over something. Or you thought you had control over something. But it’s not quite everything. You can’t quite get it done because, “Those horrible people above won’t let me do it, but when I’m a layer above, I’ll be able to fix them.” Then you find you need to be another layer above. The belief is that somewhere, at the top of this pyramid, is somebody who can actually control everything. What happens when you get there is that of course you can’t.

When you bump into the world of complexity, as Donella described it, you encounter a sense of irreducible uncertainty. Instead of trying to resolve those differences, it then becomes a question of, “How can I work constructively with them?” How do I work with the underlying form, and practice and experience and dynamics of what I’m in, in order that I might be more effective? It’s that moment where you ‘grump’ and say, “I wish it would all just make a little more sense.” And then there are moments when you’re much more lucid, and you say, “Ah, well it wouldn’t, would it?” Does that make sense?

d4e: Applying complex systems is such an ‘evolutionary’ endeavor. It’s built upon the work of an entire network, but it continues to evolve and in practice, it’s inseparable from its environmental context

JA: And that’s it. But of course, hey, there is only one system, and it’s the universe. So once you start working with a health system as a system, or a financial services company as a system, or a place as a system, there is no end to it. It just keeps going. That’s part of the problem with working this way. Your work can spiral out of bounds.

d4e: Any advice on keeping work in a complex ecosystem within manageable boundaries?

JA: The key is connecting the system to more of itself. That’s one of Myron Rogers’ five maxims. Another is “Start anywhere. But follow it everywhere.” If you’re dealing with an interrelated web of things, it doesn’t matter where in that web you begin. So long as you follow it, you’ll end up where you need to be. Say you’re trying to reduce waste in a manufacturing environment. You’ve got a good sense of where to begin. The thing is not to then box it within a project initiation document and a definition that says, “We’re doing this piece of work.”

Once you start to go into it, you actually find there may be historical cultural patterns within the business, where people just don’t do certain things that they should do. There may be problems with their IT system, payment system, contracts…All these sorts of things are probably facets of the problem, and if you just stay bound within that first narrow definition of the problem, you produce another fix that keeps it working for a little while and then unravels. Great work for consultants, producing fixes all the time. But it doesn’t help the system.

d4e: True or false: “There are no best practices.”

JA: I think there are better practices and worse practices. A best practice is almost certainly arrived at through a retrospective process that says, “Let’s look at what happened elsewhere” — in other words in the past — and formalize that into a way of doing things.

There are some problems with that and some advantages. One, it’s a really quick shortcut if you’re struggling to get something done. And, the environment that they did it in is probably not the same as the environment you’re doing it in, if even subtly so. And, environments change. Usually, best practices are determined by looking back at our work and saying “Weren’t we clever?” when we may have been lucky. Probably we’re lucky more than we acknowledge we are, and therefore the best practice is probably not a true story of what we really did.

Most management books will, for instance, research what people have done on other businesses that are successful, or turnarounds that they did in historically difficult areas. These are only part of the story. I think therefore in terms of ‘best’ practice, a practice is a way of working with the world that makes sense to you. In the end, you’ve got to do it, not the person who wrote the standard operating procedure.

A big part of working with complexity is setting up learning processes that run alongside the work you’re doing, so you have a way of making sense of what you see, albeit accepting the fact that you’re always self-referencing.

d4e: How do you do that?

All sorts of ways. Emma has done learning histories, where she repeatedly interviews people throughout the process. When you get to the end of the process, people will tell you the story of how it looks then, which isn’t all the doubts, anxiety, fears and excitement and enthusiasm that they have at the various times running through it. (You can find these case studies on Heart of the Art.) When you look back at it, it helps show you how you were thinking at that time and what it was like to be at that place. It reminds you that it’s usually messy getting there. When you’re trying to do other change processes, you start to recognize that moment when it gets real. You read the management book and it all worked smoothly: You formed the guardian coalition, you created a sense of urgency, you build a vision, etc. You try and do that stuff, and people don’t like your vision, or they don’t feel any urgency — because although you can see the company’s going down the pan, they just think so what — I’ll get a job somewhere else.

You have to remind yourself that it took work to do those things. You can use blogs to do it, or I’ve done it in retail environments and trading floors as a sort of standing huddle at the end of it. A five-minute “what-happened-today.” What did we do well? What did we learn that we’re going to change tomorrow as a result?

The US military introduced something called after-action reviews, that Myron worked on, where you do that on a big scale. One of the key bits to it is recognizing that information has no hierarchy. What I see as the store manager in Walmart, what you see as the logistics director for the region, and what the VP in Europe sees may be different things — but what you see is what you see

What happens if we’re not good at sourcing information is that actually what the vice president in Europe saw is right, because she’s the boss. The key to getting to the real narrative is learning outside of hierarchy, so that what everybody sees is relevant.

d4e: So you’re getting more of an accurate view of the network by getting the story from multiple perspectives.

JA: Exactly. If you’re the VP of Europe, you can see what’s happening in France, in Germany, in England. You can say, “I noticed this pattern.” If you’re in England you can’t see that pattern, but what you see is still true.

The narrative is always incomplete. How do we make sense of it? Through our previous experience of the world. Back to the Big Five on the Heart of the Art — this is the self-referencing part. We want to make it fit what we already know, which is one of the reasons why change in complex systems is, at one level, hard to instigate. Because the new stuff doesn’t fit our existing view of the world.

d4e: How do you put people at ease in a world obsessed with big data?

JA: Do you think they need to be at ease?

d4e: I suppose that many of us like to hear that there’s a plan mapped out using best practices.

JA: That’s a very linear view of the world, isn’t it? There’s a basic rule of uncertainty that says that whenever we measure something, we lose more than we gain. But one of the things that helps is to look at what you’ve done elsewhere. Another is that you ask them to trust what they’re hearing now about what’s going on with their business and the stories that emerge from it. Does that seem more real than, “Let’s initiate the project now, these are the milestones, these are the outputs you will get?”

Here’s an example. I was working with one of the regions of England on a five-year strategic health services plan. Somebody’s there saying, “What we need to know is the implementation plan for the next five years. When are we closing this, when are we opening that, what are the spends, when does all that happen?”

My response to that is, “Can you tell me what the UK economy’s going to be doing next week?” One thing that’s happening is that we’re voting on whether we want to be in Europe or not. For us, Brexit is a really big deal. One of the possible outcomes will severely affect the British economy. So we don’t even know what the quantum of spending is going to be next year — because it will affect the tax take and how that will impact health spending, and so on. So you’re being asked to predict a gap in health spending five years forward, which is curious. You can say what you plan to do to change it and in the meantime, the entire economy could turn upside down. Is this a sensible way to do strategy? The answer for me is no.

It’s not silly to have a clear sense of what you’re trying to achieve, and the sort of quantum you’re trying to do within that. The way that you do that needs room for flexibility, adaptation and evolution. Once you start planning for longer timeframes, and big scale, and wide geographies, you have to have a clear sense of what you’re trying to achieve together. There’s the identity question — who’s the ‘we’ in all this? What is it that we’re trying to do together? And then once you know, you allow space for the adaptation and variation.

d4e: So you have to draw a boundary around what you want to achieve and the space you want to achieve it in.

JA: Yes, and just the drawing of that is the most important act in that process.

A traditional strategy approach would say, “Let’s take a top team off-site, lock them up in the mountains, and they’ll come back down from the mountain with a vision for the next five years. That is their vision, and it’s going to have a lot of truth in it, but it’s not held and owned and embodied by everyone else. So they don’t do things necessarily that do that unless you coerce them to.

You try and sell your vision through big corporate communication processes. You make videos. If you’re smart you put all sorts of tech into it. If you’re not doing that, you’re holding town hall meetings; in fact, those are probably better.

But what happens if you did it the other way around? What if you started a conversation about it, and asked, where do we want to be in five years? What could this business do? That would fulfill another of the Big Fivecriteria for working with living systems, which is that people own what they help create. They understand, then, what’s in scope and what’s out of scope.

If you believe you’re working in a biological system and that a human world is a living thing, then how do you work with the process of evolution and adaptation as a process of change? You can never direct a living system — you can only provoke it. So how would you work with an organization as if it were a living thing? The answer is you would try to grow it. You would nurture it and tend to it, rather than over-direct it.

d4e: Just like a garden?

JA: I use the gardener metaphor quite a bit. I see leaders as gardeners. You can plant stuff — you can go and invest in a manufacturing plant in Brazil. That’s planting, isn’t it? You can prune stuff — in order to make it grow stronger you have to trim some bits off. You can fertilize stuff — what would investing in it look like? Simple living thing, a garden, in comparison to an organization. But if you think about it it probably feels much harder. You might look at a garden and think, “How would I make that do anything?”

d4e: You’re not saying to abandon strategy, though.

JA: No, not at all. There’s a school of thought that says that the whole thing is chaos. And there is chaos in it. So…whatever. That’s the irreducible uncertainty piece. If you think of Lorenz — a butterfly flaps its wings somewhere and there’s a tornado on the other side of the world — absolutely, statistically that could happen. But most likely it won’t. So we don’t need to go around exterminating butterflies.

What happens at points of pivotal and monumental change is that lots of disturbances tend to line up in the same direction. We can do things that encourage that to happen, but we can’t direct that it will.

d4e: Are you talking about designing an attractor?

JA: Yeah. And what would make it ‘attractive’? How would you know? You’ve probably got to sense the existing system and its environment to do it. One of my criticisms, perhaps, of the way the health services is doing its process at the moment, is it’s looking very inward. There are lots of people living longer, lots of co-morbidity; what do we do? They recognize also that there needs to be a shift on the scale of the population. Actually the thing that will change this problem and make any solution more or less viable is probably more about the population’s view on its own health.

But those conversations aren’t taking place. We’re just looking at “how do we run the existing system?” to a far greater extent than we’re looking at how it exists within its environment.

Instead of working on identity, information and relationships, we tend to work on structures and policies and procedures. As if, if we designed our organizational form again, we might finally get it right.

The self-referencing part of the Big Five characteristics of living systems is that the system often acts to maintain us as we are.

d4e: What do you say to a brand new consultant who wants to apply the Big Five of living systems to an organizational problem to create change?

JA: As a new consultant you don’t get to do much. You get hired by a big firm that tells you to run spreadsheets and crunch data and tell the grown-ups why they’re wrong, which is difficult for all concerned. As you get allowed to play a little bit more, you can approach things differently.

I set up a program called Total Place that you’ll find on Heart of the Art, which was a whole system intervention across England. In 2008 the financial crash was happening around us. We knew there was going to be whole-scale change in the way the state engaged with the citizen. Things that we’d taken for granted or previously spent money on, we wouldn’t be able to do.

This was set up as a process of complex change, and it was set up on Myron’s Five Maxims, of which I’ve given you a few already, and that was a template for determining what we should do. The first thing is that real change takes place in real work. The people who do the work therefore do the change. So unless you engage with what’s actually happening and are specific and precise about doing that, you always get an approximation. Setting policy is not the same as doing change.

Myron’s Maxims
Real change takes place in real work
The people who do the work do the change
People own what they create
Keep connecting the system to more of itself
Start anywhere, but follow it everywhere

I used these as a set of design principles to engage in a system that found 20 billion pounds of savings.

In terms of design, if you believe that people own what they create, you can come to them with a proposition but you have to design the work with them. Don’t come to them and say, “This is the process.” You have to engage the people who are doing that work in the design, and work with them. I’ve worked night shifts in paper mills and manufacturing plants and all sorts of places, because that’s how you understand what’s actually happening in a system. It’s all there, if you go and sit with people and talk to them and connect them with different bits of this. Sometimes you end up being the center of the web connecting all the pieces, and that’s not what they need. They need their own web.

In biological terms one of the characteristics of living systems is that they tend to kill the change agent. They will take you in, digest you, use what they want of it, infect the rest of the system (infection is a great way of looking at how you do change), and then they’ll spit out the bits they don’t want. So you’d better have an exit strategy.

That’s true if you’re a consultant, and it’s true if you’re a chief executive. Which is why, I think, we see chief executives moving every two to three years. The people I respect as CEOs have been in their positions at least five to six years, because they’ve had to actually deal with the consequences of what they’re doing.

d4e: How much does it matter that people know that you’re pulling from the principles of biological systems to design organizational change?

JA: I think it matters quite a lot. You do it with people, and it’s a co-created thing. At times it can be unhelpful to get too far into the extremes of it, but I try to be very explicit about the things I’m doing and why — and why certain I think certain things will work and others won’t. I think it helps to be very open about it, because if you’re working in a systemic way, what you’re trying to do is leave the system with an enduring capacity for change.

Biological systems have that. In a biological system you adapt or die. But you adapt. It’s not a directed process.


Stacy Hale is Founding Editor of design4emergence and a generally curious person. She believes in teamwork as an evolutionary force.