Wednesday, June 18, 2008

Essential systems thinking for managers

This could be entitled "Lessons from teaching system dynamics."

As you may recall, I taught a graduate course in system dynamics at the University of Washington last winter, and I'm scheduled to repeat that course next winter.

While I won't write about that class nor about the systems thinking class I'm co-teaching at Bainbridge Graduate Institute, I wanted to note a few things I have come to think may be at the core of "systems thinking," at least in the system dynamics sense, as a result of thinking heavily about this in the process of teaching others:


  • It's very important for people to understand that many of our problems are caused by the systems we create, not by externally-imposed actions. Understanding feedback ideas seems central to grokking that concept. Incidentally, grokking that concept should bring a bit of humility to each of us, and it also opens up possibilities: if we create our own problems, then we have some control over fixing those problems. That's a lesson worth remembering.

  • Causal loop diagrams (CLDs) can be a key tool for making sense of the systems that create the problems we see. Better than that, their dialect, influence diagrams, as described by Geoff Coyle and others (for example, see SYSTEM DYNAMICS MODELLING: A PRACTICAL APPROACH), can be even more insightful and can replace stock and flow diagrams.

  • While I'm not a fan of the "systems archetypes" you may have seen in The Fifth Discipline: The Art & Practice of The Learning Organization, there's a lot to be said for understanding what structures and what behavioral patterns are tied together. I don't mean understanding as in textbook knowledge; I mean understanding as in seeing the one and viscerally understanding the presence of the other. Can you look at data (for example, business results), find the pattern that contains the data, and use that to find the structure that likely caused the data? Can you see how to change the structure to make the data be like you wish? If you have the structure in front of you, can you work the other way to estimate the type of behavior you might expect to see? Can you test your theories through simulation or through comparison with real-life situations?

  • Recognizing the difference between stocks and flows is every bit as important as I've written about before. Add to that being able to calculate mentally the changes you'll see in a stock, given the changes you're creating in a flow (mental integration, in other words), and you'll have a better understanding of the impact your actions may or may not have.


I came to the conclusion that understanding these ideas well would go a long ways towards helping anyone think more insightfully about tough challenges. Getting there isn't trivial: it seems to require a bit of math and a bit of simulation experience, and having a lot of both seem to help. By themselves, though, math and simulation aren't sufficient; this requires a lot of thinking.

What I'm coming to realize is that having people with these (and other) skills in the management meetings of our companies would be a great help. Instead of just arguing points based on best (unaided) intuition, someone might look at the data and draw some provisional inferences. Someone might think seriously about the type of structure that might have created that data pattern and look for evidence of it in the company and its environment, sketching and discarding diagrams as they go. People might understand the likely effect of interventions based on the current and proposed structure.

Many of you who read this may be able to do all of this already, and that's great! Is it helping you in your work? I hope so; it should.

Others of you may understand when you see others do this but not be able to do it yourself. That's okay; that's why Facilitated Systems is here. If you want to discuss whether your organization might benefit from these ideas, contact me today.

Labels: , ,

0 Comments:

Post a Comment

<< Home