Freitag, 17. April 2015

Bonn Agile Meetup - April 2015: Cynefin Framework

For our last meetup, I suggested we take a closer look at the Cynefin Framework. Perhaps you've heard about it, or maybe not, but this seems to be one of the big up and coming terms in the agile world, and it is also present in other communities as well (project management, DevOps, etc).

If you want to learn more about what the framework is, have a look at the resources below (and the resources they again link to). For this post, it suffices to say that the creator of the framework Dave Snowden, and I think it is a very generic, perspective/world-view framework, rather than one that tells you what do exactly.

I personally grew really curious when I read some agile coaches discussing the Cynefin course on a mailing list some months ago. So I set myself the goal of holding a brief intro to the framework at our own meetup, and started consuming as much info on Cynefin as I could:
And then I found some more:
And more by Liz Keogh:



So, after having heard through some of the podcast episodes many times (I needed to hear through them repeatedly before I started to grok the discussion), I put together an adaptation of a workshop I found described the webpages of Cognitive Edge, which is Snowden's company. You can sign up for free there to get at some of the resources. Unfortunately, they re-implemented their website right after the meetup took place, and now I can't log in there for some reason.

The workshop was pretty straight forward. We met in a bar and...

  • We split into 3 tables, each of which were going to roleplay being the managers of a small company. 
  • I set some context for the company (what kind of employees, what they produce, etc), but left most up to the imagination of the participants.
  • Each team of managers were instructed to discuss incoming "problems" they had to deal with, by placing them in the Cynefin framework's categories.
  • I started feeding the teams one by one, giving them a few minutes to discuss each one, before I gave them a new problem. I had prepared these problems up front, but as I didn't have any reference from Cognitive Edge, I just made some up - not really sure if they were well suited to the task. Some examples:
    • The company has been selected for a tax audit
    • Our product needs to get certified
    • The canteen food is horrible
    • A customer nearly died while using our product
    • Average employee sick days are "too high" (more than 10 a year)
    • We're introducing SAP in our production/ordering pipeline
  • The teams were told not to solve the problems, merely to place them in the Cynefin framework.
  • Each problem was noted down on a post it, placed on the table, and were decorated with two arrows:
    • red arrow -> which direction the problem is headed if we do nothing
    • green arrow- > which direction we want to manage the problem
  • After all the problems had been placed, we swapped around positions to hear how the other teams had categorized their own problems for some more discussion.

There wasn't really any correct answers for any of these, but I found it really interesting, and I guess that is one of the key goals of Cynefin, how the managers had to separate complicated problems from complex ones. It was also interesting to see how trying to see the problems from the various perspectives, different solutions would pop into my head.

The minus about using this fictional company, instead of using it in a real company, internally, is that the context was somewhat artificial. I hope nonetheless that we got some valuable practice in how to make use of the framework when we go back to our day-jobs.

I certainly have started seeing more things now in light of the Cynefin perspective, not so much in my job, but more when I look at why large companies and governments are having problems with their massive IT projects. I also see the same trend in education. It seems society as a whole is trying way to much to treat things as if they were in the complicated domain, when actually they should be dealt with as complex issues.

The other key learning for me was that an agile process like Scrum, operates by iterating on features moving us from complex into more ordered systems (by implementing "constraining" software). But sometimes, we actually want to shake things up a little, and move existing systems out of the ordered space, to see what things can be improved upon, and we don't really have a technique for that in our agile methods. All in all, I think Cynefin is a fresh breath of air into the agile community, which has been pretty stagnant the last years. Perhaps we could need a bit of chaos and complexity in our methods.

I also hope that we can get some more knowledgeable Cynefin coach to visit us some time, and give us a proper talk/workshop on the subject. If you know one, or if you are one, let me know!