Cynefin vs Wardley Maps - a subjective point of view

This week, I had a wonderful opportunity to attend a joint masterclass run by Simon Wardley and Dave Snowden. The first day consisted of both masters presenting their frameworks, and the second day was about answering audience questions, which mostly revolved around points common to the both frameworks.

My impression:
Cynefin operates in completely different space than mapping. Mapping is all about components (which are incarnation of capital), and Cynefin is about understanding the right approach.

For each Cynefin domain (with the exception of chaotic), a map can be created representing an analysed system. The map, however, depending on the Cynefin domain, will have different properties:

  • in the obvious space, it will be easy to create and analyse a map
  • in complicated, the map will be easy to create, and most of the time, manipulations of the map will yield in expected results, albeit, from time to time, unexpected consequences may appear (such as second order effects)
  • in the complex space, a map is difficult to create (some dependencies may not be known), and any actions will return an unexpected action.
  • chaotic space bears no maps, as we do not have sufficient understanding of the domain to
    create one.

On the other hand, any component on a Wardley Map may be in different Cynefin domain. Each time a new approach to satisfying a user need emerges, it is first Chaotic, then Complex, then… you get it, while staying in the same place on the Evolution axis.

For this reason, in Cynefin, it is possible to go backwards in domains. If you f.e. stress people well enough, they disconnect from how things are built, and begin experimenting with new approaches. In Cynefin, it is going from Complicated to Complex, but in mapping… it might be an attempt to increase the level of Evolution or exploration of higher order system or bundling, etc.

There is huge a lot of potential for combining those two frameworks, and hence my expectation is that over time, we will see more of Cynefin practices adopted in mapping, and more of mapping practices adopted in Cynefin.

Interesting times ahead!


Much appreciated @chris.daniel! Thank you for these goodies!

Interesting. I was under the impression that it mostly lined up rather neatly as Chaotic = Genesis, Complex = Custom, Complicated = Product, Obvious = Commodity…?

Mostly going off of Simon’s descriptions of the m.o.'s/behaviors in each evolutionary step, of the “pioneers”, “settlers”, and “town planners” as he calls them.

That impression is very justified, that was my way of thinking, too, unless I found a counterexample - a traditional car to electric car transition (product to product).

Note: we are not yet talking about car-as-a-service as this would be utility and completely different can of worms.

In terms of Evolution, the contract (and I think we can say cars have an abstraction around them) stays in Product. The actual implementation is also in Product, albeit the Cynefin domain moves, at least on the contract level, from Obvious to Complex/Complicated.

A small improvement in efficiency (Evolution) causes temporary change of perception. This was very difficult to express with mapping alone.

While it is possible to think that Utility maps to Obvious most of the time, Obvious can appear in f.e. Custom-built, too, if only nobody knows how to make a product, or has no desire to build one.