Cynefin is a Welsh word that roughly translates as “the place of our being”. It is the term Dave Snowden uses to describe a way of understanding and responding to complexity in a domain. He presents the following diagram to show the different aspects of complexity:
Each of the domains in the model has characteristics that indicate ways to work or to respond when working in that state. Snowden points out that this is not a two-dimensional consulting model but is a sense-making tool used to determine the most appropriate type of response to take when acting in a specific domain.
We use the model to select the approach to use to solve a problem or to take advantage of an opportunity in a business setting.
Let’s look at the first four domains, starting from the lower right corner.
Obvious: The nature of the problem is clear to all concerned and how to solve the problem is well understood. Obvious problems are ones that have been solved many times before and there are genuine best-practice approaches to use to solve the problem. It may not be easy, but there are rules to follow, and when they are followed, the problem will be solved. Activities such as civil engineering or construction (e.g., building all the houses in a new subdivision) often fall into the Obvious domain. Plan the work in detail then follow the plan precisely.
Complicated: The problem is relatively clear to all concerned and there is a well-understood outcome to aim for, but how to get there is not obvious. In this domain, the key is to identify the combination of good practices that align with achieving the outcome. There may be a need to learn and adapt along the path, but the intended result is clear to all concerned. Building computer systems to address known business problems falls into this category: there is a wide range of possible approaches that could be taken, there will be a lot of learning as the solution evolves, and the link between the solution and the problem is relatively clear. Agile development methods with their focus on strong technical practices and rapid feedback will often be appropriate in this domain.
Complex: The relationship between problem/opportunity and solution is not clear and there is a need to experiment to uncover the actual problem. You can undertake many parallel experiments, and there is a need to run those experiments rapidly and at low cost. The ideas of lean startup (build-measure-learn) are well suited to this domain. Run a small experiment, check the results, and either amplify the solution with the next experiment or pivot and run an experiment with a different focus. The intent is to experiment and learn in order to clarify the underlying problem/opportunity (in which case the problem moves to the Complicated domain and the experimentation can give way to more structured learning) or to come to the realisation that the problem/opportunity should be abandoned.
Chaotic: There is no discernible relationship between cause and effect, between problem and solution. In this state, there is no clear vision of what you need to do to address the problem, and the solutions will not be evident until after they emerge. It is important to act quickly as chaos states will devolve into more and more chaos, but exactly what actions will solve the problem are not clearly discernible. Firefighting is an example of a chaotic domain — doing nothing is not an option but exactly what needs to be done is often not directly visible to all concerned, and often the act of addressing the problem will cause other problems (e.g., water damage in parts of a building that did not burn).Kurtz and Snowden provide an example of the impact of trying to apply ordered thinking in the Chaotic or Complex domains:
A group of West Point graduates were asked to manage the playtime of a kindergarten as a final year assignment. The cruel thing is that they were given time to prepare. They planned; they rationally identified objectives; they determined backup and response plans. They then tried to “order” children’s play based on rational design principles, and, in consequence, achieved chaos. They then observed what teachers do. Experienced teachers allow a degree of freedom at the start of the session, then intervene to stabilize desirable patterns and destabilize undesirable ones; and, when they are very clever, they seed the space so that the patterns they want are more likely to emerge.A fifth domain, Disorder, sits in the centre of the model and is the state that any system must go through when moving from one domain to another.
The cliff-face symbology between Obvious and Chaotic is intended to indicate the risk (perhaps, the likelihood) that Obvious domains can quickly become Chaotic if the rules by which they are working become invalid due to some external or internal influence.
In addition to the five domains, the model has what Snowden calls “liminal” areas on the boundary of Complicated and Complex and the boundary of Complex and Chaotic. In the liminal spaces, there is lots of learning and experimentation to do to identify the best approach to tackling the problem.
In the Complex/Complicated liminal area, Snowden recommends experimentation using prototypes and iterations. For the Complex/Chaotic liminal area, he talks about the need to abandon constraints to allow novelty to emerge.
This reference material was originally published in #noprojects. It is republished here with permission.
- Brougham, Greg. The Cynefin Mini-Book. InfoQ, September 19, 2015. https://www.infoq.com/minibooks/cynefin-mini-book
- Kurtz, C. F. and D. J. Snowden. “The new dynamics of strategy: Sense-making in a complex and complicated world.” IBM Systems Journal 42, no. 3 (2003). doi: 10.1147/sj.423.0462
- Snowden, Dave. “Liminal Cynefin: The final cut?” Cognitive Edge. October 27, 2017. Accessed May 13, 2018. http://cognitive-edge.com/blog/liminal-cynefin-the-final-cut/