Radical Self-Organization: Why Should You Care?
When I describe Lunar Logic in public presentations, I often use a description along the lines of “probably the most unusual organization you’ve heard of”. We tend to do many things differently. When we do so we sometimes get pretty extreme. One concept that we brought to such extent is self-organization.
We know self-organizing teams from the context of Agile methods. The basic premise is that teams decide how to organize and manage work that they are doing. The concept served us well, both locally at Lunar Logic and generally in the industry. Its impact is limited, though, because of the underlying constraint—the application on a team level only.
Radical Autonomy
Given how well self-organization was working for us, we evolved the model to what we call Radical Self-Organization or Radical Autonomy. The basic idea is simple. We scale the approach up and give everyone equal power in terms of how the whole company is managed. Imagine that, as a developer, you have as much say in how you write code and what is on the payroll. You get the idea.
Everyone at Lunar Logic, interns included, can make any decision. Yes, that may mean spending a significant amount of company money, giving someone (including themselves) a raise, etc. The basic pattern is that anyone can make any decision.
It means that we don’t have managers. The common understanding of management is that it’s a role with some designated power where decisions are made. Since we distributed decision-making entirely we don’t need managers. If not the fact that it’s a legal requirement to have a CEO we would unlikely have one.
Decision Making: The Common Way
As you’re reading these lines you may put yourself in shoes of our potential customer and ask the question: why should I care? Ultimately all that stuff is aimed internally at the company. What’s the gain, if any, for a client?
The gain isn’t obvious yet hard to overrate once you grasp the idea. Let’s first look at how decisions are made in a typical organization. When an employee needs a decision to be made they go to a manager, have a brief conversation, then the manager makes the call. The employee doesn’t need an understanding of the whole context.
Of course, some understanding of the context, but only a partial one, would be useful when the employee has a desired decision outcome. For example, I’d love to get a raise and thus I want to hear “yes” from my manager so I will argue why it is a good idea. I am, in fact, discouraged to see the whole picture, e.g. budgetary constraints. It would only make it more difficult for me to rationalize the desired decision.
Making Decisions Without Managers
How does it work in our context? Since we don’t have managers everyone is allowed and encouraged to make decisions autonomously. However, we want decision makers to take into account the whole context. That’s what advisory process is designed for. Before making a decision everyone should ask two groups of people about their take: those who have expertise on the subject and those who will be affected by the decision. That’s how we provide the big picture.
Then there’s another ingredient, which is accountability. Anyone is free to make any decision but we are kept accountable for the outcome of our calls. This discourages to make decisions that can hardly be justified.
Things start getting interesting here. The justification pattern we employ very often is “what’s economically sensible”. This means we need to look at the cost of the decision and the value we expect to gain through making it. The cost part is typically rather straightforward. The trickier side of the equation is the value. Try to assess the value of a conference trip or a salary raise in monetary terms. It’s not that easy. Nonetheless, that’s what we expect of ourselves.
Answering the questions: “what’s the cost of that” and “what’s the value of that” is thus our bread and butter. It’s the basic skill everyone at Lunar gains quickly after joining the company. Such training proves to be invaluable in other contexts too.
What’s In It For Me?
Think of a situation when one of your developers decides how to build a feature that you proposed. Instead of just taking what’s specified in the description they critically look at available options and briefly consider the cost and value of each before proposing an alternative or going with the initial plan. Think of a product owner discouraging you to build a big part of an application and proposing an alternative way of validating the business hypothesis instead. Think of a team that balances their desire for craftsmanship with a deep understanding of budgetary constraints and how each one may affect the other.
There are many such decisions made by any product development team every day. The cumulative impact of these decisions is much bigger than the impact of simple coding proficiency. In fact, the latter matters little if you’re building the wrong thing. And the whole point of exploiting Radical Autonomy in product development is it significantly improves the odds of building the right thing eventually.
When we were starting our journey toward Radical Self-Organization we didn’t much think on how it would impact our customers. However, it turns out to be a unique ingredient of our offer. One that is especially important when we work at early stage products.
Considering the economic equation—cost and gain of each possible path—in everyday product-related discussions is something that we could expect of some experienced engineers. What Radical Self-Organization does is it makes the skill granted for everyone working at Lunar Logic, no matter their experience. After all, that’s how we all thrive at the company.