Dave Snowden

The complicated domain

RSS Feed

I am starting to get the process of creating domain frameworks/models down to a fine art after a year of trial and effort.  So far I have them for the complex and chaotic domains and for the last few weeks have ben puzzling out the complicated one.  Now I should say that I think over time there will be various sub-domain models for specific purposes so these are the first.  The process is fairly simple, although its taken some time to get there.

Firstly we have a disorder to order diagonal from bottom left.  The top and bottom are main boundary transitions in Cynefin, so that is straight forward.  Then I look for two extreme states on the horizontal and vertical axis at the same time as I play around with the axis labels.  This takes a bit of time; for complicated it was original The Longitude Problem and Creationists.  The former remains, the latter is now Time Wasters.   That gives me four states, marked in a pinkish red on the map.  The movement between the states in green indicates management action to move things back on to the diagonal.  The diagonal owes a very large hat tip to Max Boisot and thence to Ross Ashby by the way.

So if anyone else wants to have a go at creating one, feel free!  I reserve the right to comment of course ….

So to the complicated domain, the area of good practice, of experts or knowable knowns.  Here I have kept the vertical dimension of the complex domain model as the horizontal dimension here.   It talks about the degree of consensus and ranges from the belief of a small number of people (the cogniscenti) though orthodoxy to the point where everyone believes.  I initially called this Group Think, but that was only negative so I am now using Synchrony which has a strong link to the resilience literature.  When everything falls into step we are on the verge of collapse.

The vertical dimension indicates the degree to which the subject (probably knowledge in this domain) is strategic.  Ranging from critical, through significant to trivial.  So that allows we to identify two transitionary states:

  • Bottom Left

    Here we need to be in the mainline Cynefin dynamic, moving things to the complex to explore possibilities then increasing constraint to shift them to complicated so we can exploit them.  The constant shift between the two creates resilience.  Ideas transiting to the complex will not, and should not have general consensus.
  • Top Right


    Everything is in step and its no long strategically important (it may still be vital by the way).  This is ready to increase the constraints and impose best practice to get repeatability and consistency and the lowest possible cost.  Conversely something that is becoming less trivial (for example an increase in critical incidents in field operations) then I need to shift it down the diagonal towards the complex as synchrony may mean complacency and thence danger.

Then we get the two negative extremes together with the strategies to cope with them or prevent their emergence.

  • Bottom Right


    Everything is in step, there is little dissent; this spells danger.  It is the Longitude Problem, in which established experts refuse to see or understand that there is a paradigm shift which undermines their expertise.  We are getting a lot of this with the qualitative research community who can’t cope with the power transfer involved in self-signification.  They feel they are better placed to interpret a respondent’s story than the respondent, either through their expertise or playing with semantic analysis.  If as a leader you spot this happening (Orthodoxy-Critical) then you take immediate action to disrupt thinking.  Bring in competing experts, challenge receied wisdom, set intractable problems; gnerally make life really difficult for those who want to carry on with the old orthodoxes.   Gerstner did this in IBM when he fired a third the accountants to break entrained bureaucracy; good idea but it should have been two thirds, he under estimated the problem.  If its established and you can’t skill it, then isolate it, starve it of resource so that it withers on the wine.
  • Top Left

    Cognoscenti -Trivial

A small group of people maintain an esoteric belief.  As I said I originally called this one Creationism but then the statistics on the number of people who live in the most powerful nation on the planet who believe the word was created 6,000 years were released and I got scared.  There is a form of collective madness that might bring on the end of the world,  OK maybe not this Friday but you never know.  If you see things going in this direction (Cognoscenti-Significant) then the break the cabal, de-programme the cult followers, move them to other duties.  On the other hand if you are already there then harsh treatment to impose and enforce standards is probably the only solution.

Finally in the centre you just need to check and double check that you are not heading towards the simple domain too quickly.

So three down and one to go (possibly two).  None of them are complete yet and I expect them all the change, in particular this one over the next few weeks.  As ever comments and criticism are very welcome.  I also realised that they are domain not sub-domain models – sorry about that.