Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage Articles Cynefin 101 – Shared Context and Sense Making

Cynefin 101 – Shared Context and Sense Making

Lire ce contenu en français

Knowledge is not determined in advance of our experience; things of strategic and economic relevance are not waiting to be discovered, but are invented as we go.

Robert Chia & Robin Holt, Strategy without Design.


The Cynefin model can be used in different ways; for categorisation which is useful from a situational perspective; for contextualisation which is useful in taking a rain check point and deciding what you want to do and also from a dynamics perspective, where are here and need to move to here so this the journey that we need to consider.

In the first you have a situation and you are just using the framework to understand in which domain you are and therefore which approach act-sense-respond etc is the most appropriate. This is useful to ensure that you don’t over simplify the situation and attempt to address an unordered problem using traditional ordered techniques. This is a singular situation but the exemplar narrative outlined below can be used to help determine which domain you find yourself in.

In the second there is a need to take a rain check and work out where you are given all the issues that you are facing. You may need to consider where the company is going, what are the challenges facing a project or programme that is in flight or may be looking at starting a new initiative. This is about making sense of the wider situation and given all the challenges working out what needs to be done, acknowledging that not everything is simple.

In the last situation there is a need to move a part of an organisation or group of people in a certain direction and you are may exploit the dynamic aspects of Cynefin to help. This may involve removing or loosening the constrains to see if you can change behaviours or create novel ones. This is complex facilitated management1.

In this paper we are exploring the second of these where there is a need to make sense of situation as we have found that there are a large number of issues. This is the classics situation of where you are in danger of losing sight of the forest for the trees and need to take a check point in order to understand what needs to be done.

This is one of the most useful Cynefin practices and as it supports building shared context and purpose within an organisation. Here the data procedes the model and the model does not come before the data. This may seem confusing initially but all we are doing in practice is losing the constraints so as not to bias people views, therefore we don’t draw out the model beforehand but let the domains, and in particular the boundaries develop as part of the process. Although the Cynefin model can be used for categorisation, as noted above, one of it real values is where it is used to develop a model that reflects the context and the uniqueness of each organisation. As English philosopher Gregory Bateson said nothing exists without context.

All Cynefin’s practices are participative in nature so it engages the people of organisation directly and not via some third party which ensures that there is no disintermediation. The conclusions come out of the dialogue and they are not ‘sprung’ on people.

This makes this one of the harder practices to understand and it is best experienced so this paper will outline the approach and provide some guidance on execution. This is a lightly facilitated process, which means that the process is outlined and then the participants are left to effectively get on with it which fosters shared learning. The main thing the facilitator must ensure is that they leave the people to learn by doing and only providing as little guidance as possible so the views expressed are those of the people. Therefore it is not necessary to provide a detailed overview of Cynefin other than to provide some context. This can be combined with the practices that are outlined in the second paper to support defining and refining the actions.


The intent of this practice is to develop a shared understanding of issues that a project, programme or organisation is facing and to make sense of what is going on. The process is participative in nature which means that it is left to the participants to develop these insights and the approach supports the three types of logic. These are

  • deductive – we have a large number of examples and we believe this is the general case
  • inductive – based on a small sample we believe this is the general case
  • abductive – based on a hunch we believe this is the case

Most people are familiar with the first two types of logic as they are taught them most traditional schooling but few are familiar with the last. The first is commonly referred to as the only form of pure logic as we have gone from a to b, that is the general to the specific and can validate it. The second is the area of case based reasoning, where we go from the specific to the general. The last is what is sometimes referred to as the science of hunches and is typically how most scientific breakthroughs are made.

The term is credited to American pragmatic philosopher Charles Sanders Peirce who first introduced the term as "guessing”. Ab means back and duct comes from the latin duco which is to lead, so the word means to lead backwards. This is something that only humans are very good at and we need relevant experience to be able to make such leaps. Klien’s writing on the power of intuition explores how we use intuition in the workplace and is well worth reading2. Not everything is based on analysis which is an underlying theme in Cynefin.

Most external agents can apply deductive and inductive reasoning to a problem but abductive logic requires a deep understanding of the domain. This is why the participative nature of Cynefin is important as this leverages the knowledge and insights of the people who are part of the organisation. These people have developed this deep knowledge over time.

This also touches on tacit and explicit knowledge, as Polanyi said “we know more than we can tell” and it is the engagement in the exercise that allows this tacit knowledge to be accessed. By definition this means that any approach that attempts to extract or gain access to this knowledge by questions is limited as it makes the assumption that you know the questions. This is basically a catch 22 situation; if I knew what the questions don’t you think we would know what to do? So don’t look to over constrain the situation and if other issues arise as part of the exercise then they should also be included.

Please note that when we talk of disorder we need to acknowledge that we are really making an ontological error but it is where we are most of the time. This is an inauthentic state and the value of sense making is to develop an understanding of how we can move toward an authentic state, but this is dynamic so we need to acknowledge that we will never achieve true authenticity. This also exists within each of the Cynefin domains as the boundaries are not hard and therefore we should note that Cynefin is a dynamic sense-making framework. I am indebted to Mika Latokartano for providing the bread crumbs

Shared Context

Aside: Linear Contextualisation and Value Stream Mapping – one of the issues I have with VSM is that it looks to optimise an existing process and therefore it is focused on doing the thing right and doesn’t ask the question are we doing the right thing. As an alternative I prefer to use Linear Contextualisation where we start from the client perspective and work backwards to establish what would be appropriate. The advantage of this has over VSM as it allows us to consider changes to the process as part of the mapping process. This idea came from Mike Burrows

A shared context is developed by contextualisation; that is establishing the issues that exist in this environment and how they relate. We are using Cynefin to develop this understanding of the issues here so we are assuming that these are largely known, there may be an existing issue log or list of question that you are looking to make sense of. One point to note is that I would not advocate the use of brain storming as this overall constrains people (if you are in any doubt about this read Klein). If you don’t have time then ask people to address this as part of the preparation and to bring their list of issues/concerns with them. You could also consider running a narrative exercise (see the participatory narrative paper for the approach) to establish them as part of the exercise.

Remember that diversity and naivety are key tenets of Cynefin so it desirable to engage other parties as they bring different perspectives to bear. If you are doing this in the context of a department think about engaging people who you see as your clients and suppliers, those that depend on you and in turn those that you depend upon to ensure that you are taking a holistic view. You may want to keep this exercise internal to avoid ‘washing your dirty laundry in public’ but if this is the case then see if you can get some people who would act as surrogates for these external parties.

The ideal group size is of 7 to 123. Kline mentions that 12 is as large a group you want as this is the size of group in which people are still comfortable expressing themselves. Similarly we don’t want a lot of small groups as you want to encourage diversity so don’t look to have groups any smaller than 7 as a general rule. If the group is large therefore consider splitting it in a few smaller groups. You don’t need to be too concerned about ensuring the diversity within the groups as this will come out when you review the different group’s interpretation.

There are three approaches that you can use to support contextualisation – four tables, four corners or linear; of which, four corners is the method that I prefer. In the four corner method you start out with a large blank piece of paper, or a wall for that matter. For a typical group of around 10 people you will get away with a piece of A0. Don’t try to create an A0 out of four separate A3, unless you have no other option as it defines some boundaries and people will subconsciously exploit these when the issues are being placed on the piece of paper.

In each of the four corners you place what is an exemplar narrative (a good example) of each of the four core domains of the Cynefin model. These can be

  • Obvious – have seen it before and you know what to do
  • Complicated – you know someone who could work it out with some analysis
  • Complex4 – are there competing hypothesis about what is happening
  • Chaos – there is no structure and everything is chaotic in nature

Remember that there is fifth domain which is called ‘disorder’ and these are items that you have not yet determined in which domains they are associated with. These should be placed in the middle for now. The outline below shows how this should look:

The process is to write the issues on pieces paper and then place them on the piece of paper one at a time based on where they sit between the exemplars. This means that there is no right or wrong position as the placement should be based on the individual’s view of where they are relative to the exemplars, for example it was not completely obvious but if we undertake a bit of analysis I’m sure we can find an answer so it fits around 2/3 of the way from the bottom to top on the right hand side. Remember if not just about how the issues fit between any two of the exemplars but between all four. You may need to remind people of this initially until they get the hang of it (watch people and only advise if you think this may be compromising the group).

We are implying an oblique practice here so although we are looking at developing a shared understanding of the issues that we face we are also managing for emergence and insights that may help establish the direction that people want to progress in5. These are the ahah moments when you realise something and a lot of the little pieces fall into place. This is why is important to engage your people as they are the ones who have the deep understanding as they are immersed in what is going on and therefore are capable of having these insights. Few consultancies will have this level of knowledge of the area our organisation operates in and certainly willn’t have a deep understanding of your organisation’s context. Since the approach is participative in nature the buy-in any ideas will be more straight forward as they have been involved in the development and refining of them. The process is basically summarised in the following steps:

Place the issues on the wall based on association with the exemplars. This is not an absolute position but based on where do you think the issue sits with regards to each of the exemplar narratives. I must stress that this based on the relative position and it do not allow people to put a sticky in the right bottom corner as it just obvious. Nothing is ever as simple as you think so there may be a few complications and therefore you may want to see it placed up towards the top indicating this.

When all the issues are placed you now want to draw the domain boundaries which are defined by where the issues have been placed. You ideally want to do this with some tape that can be moved until people are happy with the placement (remember there is a fifth domain which can hold issues that are not yet determined).

If people can’t agree if an issue sits on one side of the boundary or other then ask them to consider if there are two issues and whether they could be split into two with one placed in one domain and the other placed in the associated domain. This gives you set of contextualised issues which you can now start to work on. Remember to keep an ear open insights that come out of this exercise which are instances of abductive logic.

Each business is unique therefore the solution or solutions that have worked for other organisations may not be ones that work in this context. It is the old adage, if I have a hammer then everything looks like a nail so beware of people who have quick fixes or believe that they have an answer as it worked for them in their last role or some other context.

The next phase is the development of a series of initiatives that start to address the key issues that have been identified, embracing the insights that have come out of the exercise6. Remember that there may not be a strong justification for an idea but as long as there is a coherent argument it should be considered. This approach to development of the initiatives and refinement is covered in the second paper. It discusses the use of portfolio forms and ritualised dissent to review and formalise these initiatives.

Closing Comments

You can use these techniques when you want to consider direction to help inform strategy, starting a new project or forming a new team and want to accelerate the jellying process by developing a shared context and ensure alignment on purpose. The linear contextualisation approach is also useful if you need to review workflows and provides an alternative to the lean practices of value stream mapping.

The technique is scalable and you need not limit involvement to 20 to 30 people. It is possible to run this for whole departments or organisations and involve large groups of people which makes for some interesting discussions.

Note: these practices are the registered property of Cognitive Edge and are covered by the following user agreement - If you register as a member of the Cognitive Edge Network they can be used for personal use, or within an organisation but if they are used commercially you need a ‘rights to use licence’ which is obtain by becoming a premium membership of the network – see


Hexagon stickies are available from - - go for the small ones. These have two glue lines and are better that the one with a single line as they are more durable and stay flat when attached. Cognitive Edge also have stocks but you can also just use traditional post-it notes.


  1. Robert Chia & Robin Holt, Strategy without Design: The Silent Efficacy of Indirect Action, 2009
  2. Gary Klein, The Power of Intuition: How to Use Your Gut Feelings to Make Better Decisions at Work, 2004
  3. Nancy Kline, Time to Think: Listening to Ignite the Human Mind, 1999
  4. Michael Polanyi, The Tacit Dimension, 1966

About the Author

Greg Brougham is an experienced systems and technical architect specialising in card and payments system. He has worked on the renewal of the main UK domestic clearing and settlement system and also a similar system for one of the major card companies. It was while working on the later in the mid 0’s that he thought that there must be more effective means of delivery than the prevailing approaches that were preoccupied with utilisation. This led to a review of the prevailing management literature and an interest in Theory of Constraints, Systems Thinking, Toyota and more recently complexity (in particular Cynefin) and the cognitive sciences.


1 This leads into ABIDE (Attractors; Barriers; Identities; Dissent/Disrupt; Environment) which is the Cynefin approach to management.

2 There is also an interesting discussion on intuition in the Heath brother’s Decisive. They reference Hogart’s work and note that this is only likely to develop where there is a learning environment.

3 Jeff Sutherland in his last book maintains that this 5 to 9 but this is a different context. What he is looking at is the smallest group that has all the skills necessary to do development.

4 The original text here was “we have no idea but in hindsight it was obvious what was going on” but I noted that Dave Snowden had changed this in a comment on a blog post and thought the new text more relevant as it focuses on the now, not the past.

5 Direction is not the same as strategy. Strategy is the means that you use to move in the direction and fulfil the purpose.

6 Do make sure you retain the maps as these are useful source of other initiatives. We once concluded an exercise and were discussing the initiative when I turned around to find one of the programme managers removing and grouping the stickies from the wall. Fortunately I had taken a photo!

Rate this Article