1516091940-featured.jpeg
Dave Snowden

A sense of direction (3)

RSS Feed

I really should be cutting back on the email and task backlog at the moment but I keep getting distracted by the print of Cynefin on which I have made a growing body of notes so its probably best to get it out of the system. Tomorrow I head off for a cottage on Ynys Môn to break the back of the book over eight days. I’m staying next to the reputed site of the druids last stand against the Romans in Rhoscolyn whose name derives from Rhos meaning moor and colyn meaning column from a Roman boundary marker. I have coastal walks around Holyhead and down to Traeth Crigyll beach, not to mention one of the best gastro pubs on the island (next door) to provide inspiration and refreshment. I’m having to take non-book work with me but I’m rationing my time and will try and blog daily about the work I am doing. This series, and the Christmas one are a part of the process.

That aside lets get on with the description of how we set goals and targets by Cynefin domain. In my last post I made the important point that the traditional KPI approach only really works in the Obvious domain and to a degree in Complicated. However the main issue in Complicated is the process and validation of mechanisms of professions and the various controls. I’ll return to that in tomorrow’s summary.

 

Complex

The system is dispositional without linear material cause. I’m going to assume past knowledge here by the way and I don’t have time to make all the linkages to prior posts so you may need to do a search or two if the material is familiar.

Given the nature of a complex adaptive system have some key things we need to pay attention to:

  1. How accurate is our description of the here and now?
  2. How much of the recent past can we account for/model by the modulators we are aware of?
    The more we can account for the lower risk we have going forwards.
  3. Have we established realtime feedback loops?
  4. Have we disinter-mediated decision makers?
  5. Can we make outlier views or opinions readily visible? Will people pay attention to them?
  6. What level of diversity do we have in our networks and response mechanism?
  7. How broad a knowledge base can we draw on in a timely manner?

Now notice that all of these can be measured and to a large degree targeted. But they are all about measuring the nature of our knowledge of the eco-system not the goals. When we set targets they are about setting a direction of travel, shifting to the next stepping stone: What can I do to create more like these, fewer like those. I’ve written a lot of posts on this subject, for example on the banality of measurement but I haven’t out it all together yet.

Improvisation is critical here and the idea of bricolage, using what is too hand, is vital. We really can’t afford to be too efficient here as that will damage overall effectiveness by reducing variety in the system. Here exceptions are opportunities, in Complicated they are check points, in Obvious deviances to be eliminated.

Liminal: Complex to Complicated

As a body of knowledge starts to stabilise are ability to make statements about desired future states increases. So here we are creating a professional capability. Critically that is not just about training, it is about the process surrounding that training, peer validation and apprentice based learning. As the generations of knowledge increase these become custom and practice, but in the early stages we need to be more paranoid. Hence by strong objection to accreditation and certification schemes based on attending courses possibly enhanced by simplistic multi-choice questions.

We are shifting from a measure of the overall eco-system to ones of our ability to structure, to create predictability. Obviously we need to be very, very careful before making this transition, hence liminal.

Liminal Complex temporarily into Chaos

One of the standard questions I get when I challenge the whole idea of explicit goals is Kennedy’s famous 1962 challenge to get a man on the moon by the turn of the decade. Historically I’ve responded that there is a place of aspirational goals but we tend only to remember the ones that work not the ones that fail. I stand by that as the technique has more failures than successes. However I realised that its really a way of dipping into chaos, to challenge assumptions that things can’t happen. I’m still working on trying to define how we create legitimacy here but the principle is clear.

One aspect of this by the way is the ability to commit resources to allow the achievement of the goal and also to reallocate resources or abandon other objectives. Timing of course is key, this is a way of realising an as yet unrealised potentiality; but the potentiality is there to be realised but it may only be perceived by the few.

Chaotic

With the development of the liminal, contained, domain between Chaos and Complexity then Chaos itself is always a crisis. Either a failure of containment or a complacency induced collapse from the Obvious. Here our goal is survival and the ability to allow and permit draconian decisions to stabilise the situation is key. Exceptions now become an objective for extemporising solutions.

Our solutions and measurements are closer to Complicated than to Complex. We need to create professional approaches to crisis management. Small teams trained through multiple exercises of failure able to cope with the situation. Distributed decision support which does not require disclosure of intent thus maintaining options. I’ve got a lot of posts and links there but I need to pull a lot of this together, including the material original used in the Chaos domain model (now being replaced by liminality and that link is to an early version). More on that when I get to Crisis management in the book.

I’ll conclude tomorrow.

Top