
Brian Robertson explains how organizations can adopt decentralized self-organizing structures. He talks about the tyranny of consensus, distributed autocracy and how to channel dissent into organizational learning.
Brian Robertson is an entrepreneur, organizational pioneer, and author of the book Holacracy: The New Management System for a Rapidly Changing World. He is most well-known for his work developing Holacracy, a self-management practice for running purpose-driven, responsive companies. Brian currently works as a business partner at HolacracyOne, the organization he launched in 2007 to steward the Holacracy practice and assist other organizations seeking to adopt it.
Extinction Rebellion is a leaderless, decentralized, international and politically non-partisan movement using non-violent direct action and civil disobedience to persuade governments to act justly on the Climate and Ecological Emergency. It was established in the United Kingdom in May 2018.
Christophe Bruchansky: Extinction Rebellion have been very vocal about their support for the Holacracy philosophy, even though they don’t follow the method strictly. Have you been in touch with some of their members and do you know what response Holacracy inspired in them?
Brian Robertson: No, I haven’t been in touch with any of their members. They didn’t work with any of our certified coaches or licensees throughout the entire beginning process. They had a certified coach come in and help them a little bit more recently, but it was much later in the process. Holacracy has spread quite a bit out there, especially for anyone looking at more decentralized self-organizing structures. I’m guessing somebody from Extinction Rebellion (XR) heard about it from just that kind of organic viral spread.
What do you think about the way they adapted the Holacracy framework? Any tips for similar grassroot organizations wishing to adopt a decentralized structure?
My advice is usually don’t adapt it (laughter), at least not at first. That’s because until you really understand why it is the way it is, it’s really hard to make a change without breaking something core and what it’s all about. But it’s really hard to do that if you don’t work with a coach or get some training or somehow get the expertise internally. Sometimes people do proceed without a coach and without expertise. And sometimes that works really, really well for them. And often it fails.
XR is lucky. Pieces that they’re using seem to be really working for them. Holacracy is a meta-framework for customizing anything and everything in your organization, but if you start messing with the framework itself, it’s just too easy to break something along the way and to end up in something that really isn’t what Holacracy is going for: you tend to start sliding back to some other old power structure, whether it’s a top-down management hierarchy or just who has political clout. My recommendation is to adopt it until you understand fully why it’s there and why it is the way it is. And then by all means, if you want to start forking it and evolving it, do it, although most organizations that get there are fine, they don’t need to customize it because they can already customize everything with it.
If you can, get a good coach or get somebody to do a training session or at the very least absorb every free resource online you possibly can. There’s a lot of free resources: there’s a community of practice that will answer questions on the forums for free and there’s free software that supports it. If you do have some budget, at least try to get a little bit of training, there are fairly inexpensive training courses run by actual certified Holacracy coaches that will help you immensely. The failure rate is really high when people have no support and no training. So getting help is my number one tip.
You once said that Holacracy is a rule system for anarchy, without any ruler, but some might argue it’s at the cost of exhaustively detailed procedures. Would you recommend Holacracy to any organization wishing to get rid of their rulers, or is there a trade-off?
Let me just speak to the first part before the question. It’s definitely at the cost of structure, but I don’t think that’s a bad cost or a bad thing. Look at the exhaustively detailed procedures we see in most large bureaucratic organizations, they’re way worse than Holacracy. Imagine you went into a large organization today, a conventional one, and you tried to write down all of the rules at play for how this system worked. How many pages would that take? It would be massive. The entire Holacracy rule book fits in about 20 pages! There are some detailed procedures, but it’s not anywhere near the exhaustively detailed process we see in most conventional companies today.
It’s like a feudal system. You have Kings and barons and peasants. You don’t need exhaustively detailed procedures for how power works, because it’s really simple. The King can do whatever the hell the King wants. Do you want that, or do you want rule of law? Rule of law requires courts or some kind of dispute resolution process because you don’t want to just fall back to the King decides... The trade-off is in the complexity of adoption. I don’t think there’s a trade-off once you get there.
What I hear consistently from organizations that succeed in making the shift to Holacracy is they’d never go back. Once you’re there, it can do anything management hierarchy can do. And I think more efficiently and more effectively once you know how to use the rules of the game. We’ve grown up in top-down command hierarchies, our families, our schools, our societies, they’re all organized this way. It’s familiar and comfortable. To help people figure out how to operate in a system that is deeply empowering is really, really difficult. That’s why I say it needs a good coach or at least some training.
Adopting the Holacracy constitution is not the hard part. You can adopt it with a signature, from the CEO or whoever holds power in your organization. That doesn’t change anything until members of the organization know how to use the rules. And that’s a massive multi-year learning curve. In an organization which is highly volunteer-based, it can be even harder because often people don’t have time and energy to learn a new way of organizing. They’re volunteering for the mission of the organization and they want to go out and execute it, which is great. What I often recommend in a case like Extinction Rebellion with a lot of volunteers is to have the core organizing body use Holacracy. Then you might have the whole organization use Holacracy but you take specific steps to mitigate how long part-time volunteers need to actually learn the rules. That’s probably what XR has done by not really adopting Holacracy at the front lines. I suspect there is some Holacracy expertise in that organization, it’s just not spread uniformly throughout.
Here is a phrase from Extinction Rebellion’s open documentation: “We promote the ideas of Holacracy over consensus”. Do you agree with this assertion? What’s wrong with consensus?
Yes! One of the main goals I had with Holacracy was to get me out of what I call the tyranny of consensus, which is where everyone has a voice, but nothing gets done (laughter). When I was early on in the evolution of Holacracy, when I was looking for a better way to run my company at the time, one of the things I tried was consensus because I wanted everyone to have a voice to drive change. But what I found is when everyone has to agree on things, everything slows down. There’s wisdom in every perspective, and consensus integrates that, but it’s at a huge cost. Consensus can be a swamp. You can get stuck and deadlocked. Even when you get a decision, it can take massive amounts of time and energy.
Holacracy cuts through all that. It’s intentionally not consensus based. Holacracy is distributed autocracy. Instead of trying to come to consensus on everything, Holacracy has a process that does integrate and reconcile multiple perspectives. So it is still multi-perspectival, it gets multiple perspectives in, it gives everyone a voice. But in the governance process of Holacracy, we’re using that to get clear who leads what, what roles we need and what authority each role has to lead autonomously. So the person in the role can then go lead autonomously without needing consensus.
And yet there are also boundaries on that. There are limits to your autonomy and your autocracy in any system. And Holacracy focuses as much on defining the limits as defining the freedoms. Because if you don’t know your limits, you actually don’t know your freedom. So what Holacracy does instead of just trying to get everyone to agree on everything is really focused on distributing who controls what, within what limits and with what freedom. And then you empower people deeply to go lead their area. And if they need something from others, they know who they can go to for what, what you can count on from others. And when that needs to change, or you don’t know, you have a governance process that gives everyone a voice to evolve the structure of who makes which decisions.
When you allow multiple people to have a voice, it generally does work better, but Holacracy avoids the huge pitfall of the deadlocks, the slowdowns, and it integrates the wisdom of autocratic structures. So it’s an integration of wisdom of consensus and wisdom of autocracy. It just doesn’t organize that autocracy up a traditional command hierarchy.

How does the Holacracy framework deal with dissent? What happens if some members of an organization practicing Holacracy disagree with its ethics or the direction it’s taking?
I get asked this a lot, but the question itself reveals so much about the assumptions we have about most organizations. When we’re coming from traditional organizations, management hierarchies are stuck in the tyranny of consensus. Quite often we call meetings for everything. We discuss everything. We try to get buy-in on everything. When that’s the frame you’re coming from, dissent is a real issue. Just saying the manager is going to decide is fine theoretically but not in practice. Managers don’t do that in those management hierarchies. They wait for others to come to an agreement. Or they talk about it forever.
I think we can do better than just toss it up to a manager who’s often removed from the situation to decide. Dissent is welcomed in Holacracy. There’s lots of dissent in my company. Let me give you an example. I do a lot of public speaking. We get many invitations and I have a colleague that fills the role of booking talks for me. So she gets all the invitations, she builds relationships, she negotiates with them and figures out which conferences to send me to. And then at the end of her process, she presents it to me and says, here’s a conference. And many years ago, I would sometimes shoot down the decision. I would say, I’m not going to go. It’s not worth my time. I think it’s the wrong market, or it’s not big enough or whatever.
My colleague has autonomy to do everything that is part of her role, but I also have autonomy to say, no, I’m not going to go. And that was creating a lot of tension for us. Instead of trying to come to a consensus on that, she showed up in our governance process and proposed a new expectation on my spokesperson role. She wanted to add a responsibility on my role to publish my criteria for which talks I’ll accept to go. And she wanted the power to choose the talks, as long as she’s aligned with the criteria. And it took about two minutes in that governance meeting to get that new expectation out into my role. An interesting footnote of this story is that she was the newest hire in our company right out of college. And I’m the founder of the company. And yet it took two minutes.
So we start out with total autonomy, we’re doing our own thing, but then tension develops and tension leads us to evolve the expectations and the power structure. So what Holacracy does is to channel dissent into organizational learning. I still may not completely agree with every conference she picks. And she may not completely agree with the criteria I define and that’s okay. We don’t need to solve all of that. We have enough of it that it works. And we free each of us to go lead our area without needing complete agreement from everyone.
Here are some characteristics of postmodern organizations: they are self-reflexive, decentred, deconstructionist and non-totalizing. Would you say that Holacracy is a postmodern framework? And if yes, what would be the main reason why?
I don’t know if you’re familiar with Frédéric Laloux’s work. He wrote the book Reinventing Organizations. He uses a model that looks at evolution of value systems and cultures, and it goes through multiple stages. One of those stages is what he calls a postmodern organization. He puts Holacracy on the stage after that actually, which means that Holacracy has the power and values of the postmodern organization. But it is not that paradigm: it is something new and broader. It reconciles some of the paradoxes inherent in postmodern organization, like the value of integrating perspectives with rapid workable autocratic action, which postmodern organizations often struggle with. The same goes for deconstructionism, you also need to rely on some kind of norms and standards and there’s a place for “going with” instead of deconstructing.
So I would say that Holacracy includes the values of a postmodern organization, but I would put it in a category that is post-postmodern. That said, Frédéric Laloux may be using a slightly different definition of postmodern organization from what you’re referencing.

You said in a 2014 video that Holacracy is a “meta-game for agile organization”, that it allows organizational processes and structure to evolve constantly and adapt to anything coming up. Has the Holacracy framework itself evolved? And what is driving its evolution?
Yes, absolutely! It’s one of the things I most appreciate about Holacracy. It itself is an evolutionary system, just like the way it brings an evolutionary system into companies.
Holacracy is encoded in a constitution, the same constitution used by the thousands of organizations today doing Holacracy. Because it’s a meta rule system, it’s not telling you how to organize your specific organization. It’s giving you a framework for changing anything and everything within it. That framework itself though, is open source and it’s managed just like open source software.
I think of it like an open source operating system for an organization. So just like Linux, most people don’t get into customizing Linux. They just use it as is, and they customize within it. You can still have a lot, you can define different apps, you can install different plugins. There are so many ways to customize without going into the code of Linux. And the same is true with Holacracy. You can adopt the operating system of the Holacracy constitution and customize within it, but a very small minority of real Holacracy nerds, just like real Linux nerds, will actually submit changes. And there’s a whole peer review process for changing the rules of Holacracy itself. There’s an open source development process. You can see this on github. There’s an issue database and it’s version-controlled. The current released version of Holacracy is 4.1, version 5.0 is in our second beta and it should be released in a couple of months.
What’s driving that evolution are real issues and real organizations. Holacracy does not evolve because somebody had a good idea or a theory. It evolves because there are people using Holacracy. They run into something that the rules don’t elegantly handle yet, where they have a need that the rules don’t support. And then they feed that back into the development process. And Holacracy evolves to accommodate that.
Sometimes they find a rule that’s overly restrictive. We’ve seen this in Holacracy 5.0, we’ve removed some of the rules that we realized were getting in the way, so we kind of carved them back. And in some other cases, there are rules that are a little more restrictive because we found that the lack of something was actually getting in the way. But it’s going mostly the other direction with this version.
It’s just like open source software, real users with real edge cases are submitting feedback. My organization is heavily involved. Often a user will just submit an issue, but they don’t want to get involved in actually trying to change it. And then we’ll take that up. We’ll investigate it. And if it makes sense to us, we’ll then start writing a change to submit to the community. It is first and foremost a community process.