Subject: Re: Heresy 101

[Jim Thompson <73424.1506@compuse]

Regarding CLD vs. S&F diagrams:

Jim Hines (jimhines@interserv.com) said:
"Id like to think that a good system dynamics person is technically able
to use either form at the initial stages of a project and chooses one
form over the other based not only on personal preference but also on
the needs of the group he is working with and, importantly, on what is
salient in the problem being addressed."

Most of my work is done in workshops with large company clients, and about
two-thirds of the work is with a team of managers, helping to develop
strategy. The term, strategy, can evoke many different images, so let me
bound it a little by saying that these engagements are about finding long
term policy choices for gaining advantage over a competitor.

Most of the managers with whom I work are very accustomed to making
day-to-day decisions quickly and much less accustomed to reflecting on
policy. The decisions they make are usually aimed at short-term resource
allocations ("do this in such and such a time," "you can spend X dollars
to get that"). This decision-making activity comes naturally and follows
a cursory review of circumstances that, for the most part, is like a spot
check of reference mode data that supports a mental model.

The managers on a strategy development team are likely to be well educated
and richly experienced in the relevant industry. Accordingly, they view
themselves as expert and view the consultant as competent in a process
more so than industry experienced. When I have tried to introduce either
stock-and-flow or causal loop diagramming to such groups, the reaction has
been mixed but is quite often very negative. The managers do not wish to
learn a method that takes a lot of time to understand. On the other hand,
when weve been more "playful" and produced spaghetti diagrams, the
reaction has been largely positive. Briefly stated, these folks like
"systems thinking" at least to start. In a group setting, if I ask too
often, "why do you think this-causes-that," the reaction again can be
largely negative.

So Ive developed an approach of allowing the clients to exercise their
imaginations and sort of guide my hand as we draw in the relationships.
Later, I can go over this relationship map and ask off-line questions of
participants.

Returning to Jim Hiness comment, I would emphasize "the needs of the
group" above the methodology. The methodology can grow on folks, but that
is less important than engaging a group in discourse. System dynamics can
become a hindrance to discourse. At least for me, the goal is a logical
and well-reasoned outcome. If the group is unaware of the the tools
employed (diagramming methods, formal models, computer simulations,
etc.), it hardly matters at all.
Jim Thompson
Gemini Consulting, Inc.
73424.1506@compuserve.com