… and the butterfly stamped

This method is a group or workshop based lightly facilitated exercise that creates awareness and understanding of the Cynefin Framework, in a hands on practical way. It uses a set of data points (in Contextualisation methods, these are typically narrative fragments from the actual culture of the host organization or target group) and groups interactively discuss the placement of these onto each Cynefin domain. Many of these data points may be ambiguous or similar. This promotes discussion and group sensemaking. Typically this Method is used as a introductory exercise and a learning and awareness approach, before working with real narrative fragments. Historically one of the data items in the original workshop, ‘butterfly migration’, reminded one of the Kipling short story ‘The Butterfly that Stamped’. The story itself is interesting because it illustrates the reality of the existence of multiple perspectives and the strategic advantage that can be gained if one manages those perspectives well. This exercise is essentially the same as Cynefin categorisation, except that the items being placed have been carefully chosen by the facilitator.

Typical Uses

  • Familiarising organisation / participants with concepts of complexity and the Cynefin Framework



The main decision is what set of data to use. Depending on how much time you can spend on this (and how much help you think people need with grasping the concepts underlying Cynefin) you can include anywhere from 20 to 50 items. It’s a good idea to include more items than you think people can place in the time you give them, to put a little pressure on. Also, if there are lots of items, slower-moving groups don’t need to place all the items in the time given, but faster moving groups will have more to work with.

There are three choices when it comes to procuring a list of items:

– Use the generic set from Cognitive Edge
– Alter the generic set
– Create your own set as described above

Finally determine what groups will conduct the exercise. If you are using this as a preliminary to the four tables method then it makes sense to start with four groups of at least four people each to allow for easier follow through

Things You'll Need

  • A dataset printed on sticky labels
  • Butcher paper
  • Hexagonal post-it notepads
  • Tables or walls on which the butcher paper can be placed and people can work
  • Introduction to the Cynefin framework; this video may be used


http://cognitive-edge.com/wp-content/uploads/2015/01/CEPF-Toronto6Mar2012Butterfly1.R2-wpcf_300x400.jpg http://cognitive-edge.com/wp-content/uploads/2015/01/CEPF-Toronto6Mar2012Butterfly2.R1-wpcf_300x400.jpg http://cognitive-edge.com/wp-content/uploads/2015/01/CEPF-Toronto6Mar2012Butterfly7.R1-wpcf_300x214.jpg http://cognitive-edge.com/wp-content/uploads/2015/01/CEPF-Toronto6Mar2012Butterfly8.R1-wpcf_300x214.jpg


Task Comment
Explain the Cynefin framework and ensure the you make the point that Cynefin is a sense-making framework, in which the data precedes the framework. Contrast this with a categorisation framework in which the framework precedes the data. You don’t need to worry about over emphasising the data precedes framework point, most people will simply draw the framework any way. Its a good illustration of pattern entrainment. The introduction can be supported by the example of the magic roundabout against traffic lights; the children’s party story or the outline linked in the overview. 
Now give people the list of items by handing out the pages if sticky labels and also a pack of hexes (use a different colour for each table). Tell them to put each sticky label onto a hexi and then tell them to recreate the framework using the butcher paper and the labels. They should do this as a group, so get the size of the group right before you start the exercise. Resist the temptation to give them any more help or to remove the ambiguity from the instruction. Under no circumstances encourage them the draw the framework and distribute the items. State that people should feel free to disagree and resolve issues in whatever way they think is best.Remember it is not your place to decide if they were right or wrong, or to form conclusions based on what they do or do not do. Stay out of it, let them get on with it.


When finished tell each group to appoint a spokesperson who stays with their word product then rotate the rest of the group around the different tables asking them to identify:

– What is the same
– What is different
– What most surprised them
– Then engage all the groups in a conversations.


Once this is complete you may find it useful to explain how the list was constructed. You should also point out that if the group started by drawing the framework then they failed to pay attention to the earlier description of a sense-making framework. One of the points of this exercise is that there is a pattern, but because people start with the model they have, and assume its a categorisation one, they fail to see the more basic pattern of five items in each subject group.

Do's and Dont's

  • It's helpful to mention that the items were specifically chosen to be difficult to place because of the existence of multiple perspectives, and that people should feel free to disagree and resolve issues in whatever way they think is best. Note that you are specifically not telling them how to resolve issues; that is not the purpose of this activity, because there are many possible ways one could do that (split items, copy items, draw shapes, draw connections, etc etc), and you want people to think on their own about how placing items in Cynefin space works. You may need to mention that there are no "correct" answers and that the exercise is not a test.
  • Some behaviours (Cynefin-specific or otherwise) to be aware of: You might notice that people are rarely placing any items in the space of disorder. This might mean that people are too focused on nailing down answers and not open enough to uncertainty. Or it just might mean that you didn't explain that part of the framework well enough. Conversely, you might notice that people are placing too many items in the disorder space. This may indicate a reluctance to engage in debate about the nature of items and/or a lack of energy and enthusiasm for the task (and thus possibly for the tasks ahead as well). You might notice that people are avoiding placing items in certain (non-disorder) domains - the usual pattern is to avoid the chaotic domain. This is often for lack of understanding of what that domain means; for example, a belief that chaos only means absolute mayhem and destruction may lead people to believe that nothing short of apocalypse belongs there. You might notice that people are resolving all disagreements about the placement of items by caveat, perhaps by the most opinionated or highest-ranking person. That will tell you that you need to make a special effort to get people to debate and discuss in the remainder of the workshop. You may also need to pull some people out of the groups for a concurrent exercise later on. You might notice that people are rarely if ever taking any steps to resolve conflicts about placement of items by doing anything to the items - for example, by splitting them into multiple items describing different aspects of the original items. Like placing too many items in disorder, this may indicate a willingness to accept a quick answer and a lack of interest in or enthusiasm for the process. It may also indicate that people have difficultly accepting that there can be more than one "answer". You might notice that people have placed anything outside of their main area of focus in complex or chaotic space: the biological items, for example, or the societal ones. (This is why it's best to have items from several domains including unfamiliar ones in the list.) When you see this happening it means that people are not considering the similarities between different types of things acting in complex ways, which is an abstraction they need to be able to perform later when they do more work with Cynefin. You might want to give a few examples of ordered and unordered systems behavior in other domains (termites building a mound, magnets aligning, traffic jams) and show how they are similar to examples in domains the people are familiar with.