“Self-organization” is a suitcase word — one which everyone is defined in many very different ways. Does self-organization even work? and will they do what needs to be done? are just two of the many questions I regularly hear as a consultant about this loaded term. Team leaders, managers and product/business owners have all thought about the mythical magic that self-organized teams are able to achieve but they cannot quite bring themselves to believe in it. More commonly, they fear that less control will ultimately lead to organizational chaos.
What is self-organization?
Have you always been able to predict whether you can work with someone or not? There is a gut feeling, for sure, but certainty? Human interaction, like teamwork, belongs to the category of “complex things” in life. Complexity here means “only somewhat predictable with more or less surprises”. Having said that, it sounds reasonable to me that the people involved in working together know best by experience — from the successes and failures of experimentation — who is best to work with whom, when and on what. They are best suited to decide whether someone is a good fit for the team or better placed in a different one. Let’s consider both the definition and scope of “self-organization”.
Self-organizing vs. self-directing at the team level
But what if all teams self-organize like that and it looks like a flustered beehive? Honey will likely end up where it’s not required and larvae fed with the wrong stuff — utter chaos! Although every team is buzzing on the small parts that comprise the whole, it won’t quite add up to a successful outcome.
Organizations suffer from “local optimization” because they only deliver value as a whole. This means we need a clarification: self-organization is not the same as self-directing. As with all complex concepts, you’d be surprised to see what an autonomous (self-organized and -directed) team will look like and come up with in the end. The results might be great, but they could also fail to hit the target.
The real magic of a beehive is that all bees know how to work towards securing offspring. All this happens without the Queen Bee micromanaging any bee (or, indeed, team member...).
Governing vs. enabling constraints
When you start to implement self-organized teams, you’ll sometimes find subconscious reflexes being triggered. People become concerned, asking questions like: What about quality assurance? Efficiency in ways of working? Leveraging of scale effects? Don’t we need policies and control instances?
Such reflexes aren’t bad — all of these issues are important. However, it’s necessary to funnel this energy into a certain type of constraint. Simply stifling the experimentation process to find an optimal “performing” state — which is different from team to team — by enforcing tight rules and frameworks through governing constraints - is counterintuitive.
Instead, it’s more productive to think about your approach in terms of “scaffolding” or “guardrails.” Doing so ensures you leave creative leeway for reaching the desired levels of quality, efficiency and value creation. These enabling constraints act like an organizational spine, providing structure around which practices and activities can develop as needed.
An advantage of enabling constraints is that they are meant to become obsolete. Teams don’t get clear instructions, but are incentivized to acquire the described skills and attitudes to meet organization‘s expectations. This creates managerial room to move the scaffolding elsewhere — a prerequisite for adaptivity and innovation.
Implementation advice
So, if enabling constraints can provide support and direction, how can we put them into practice? The key is Thoughtworks’ notion of sensible defaults and setting transparent strategic expectations.
Sensible defaults
Evolution means to build up on things which have proven to work. For example, sensible default practices represent a starting point, which, when applied to a certain context, gives you a head start or at least a foundation to chew on until something better is found. Or, sensible default principles state those aspects of work which absolutely cannot be compromised on. If we set a principle for code quality, that becomes a measure for “what good looks like”.
Transparent strategic expectations
Marathon runners know that visualizing the finish line keeps them motivated. Planning, memorizing and executing each single step is tedious, though. An effective strategy provides goals and hypotheses that will help you reach the end. It needs to be clearly articulated and in a reproducible format for teams to self-organize around it. At Thoughtworks, we work with principles and processes as described in the EDGE operating model, and with representation tools like the Lean Value Tree to make investment decisions and goals, value hypotheses and success metrics transparent.
In a nutshell, direction is a powerful enabler for teams. That is even more true for self-organizing teams because they are able to set themselves up for aspired outcomes and don’t depend on instructions. Improving on clarity of direction can mean to set up reasonable constraints like sensible defaults. They allow for full focus on core value creation. In addition, with guardrails in place, teams appreciate clear strategic expectations. They motivate and give purpose. Altogether, this increases the chances of a match between teams’ results and organization’s needs.
Disclaimer: The statements and opinions expressed in this article are those of the author(s) and do not necessarily reflect the positions of Thoughtworks.
