Stephen --
I like your incoherent babble regarding focusing on decisions as the key
to defining organizational structure. Your comments brought to mind the
following paper: Huber, George P. & Reuben R. McDaniel, The
Decision-Making Paradigm of Organizational Design, Management Science, 32,
5, May 1986. My feeling is that the approach put forward in this and
other papers that focus on decision-making misses an important fact about
organizations. While an organization is in some sense an information
processing and decision-making system, it is more fundamentally a
communicative social structure. By that I mean that the human members of
an organization interact with one another and it in the patterns of these
interactions that processes and structures emerge. As Terry Winograd puts
it, rather than taking the view that People process information and make
decisions, we can take the view that People act through language. Another
way of looking at decisions within an organizational structure is to view
it as a network of commitments. Individuals act in roles and have
specific (or not very specific!) accountabilities, and they are constantly
negotiating and re-negotiating their commitment through interactions with
others. By mapping out the specific interactions between specific pairs
of individuals, we can articulate processes in much more specific ways.
The problem with the People process information and make decisions view is
that is doesnt identify who one is making a decision for (i.e. who is the
customer in this interaction?). Every decision (which creates a commitment
or promise) happens in the context of a specific thread of communicative
interactions between specific individuals.
You said: If you can determine the core processes of the business... and
further identify the key decisions that must be made in those processes,
the location of the decisions should be manifest, and the nature in time
obvious as well. By nature in time, I mean are the decisions needed
before action can be taken, or are they related to a feedback loop that
can afford to assert control after the fact?
I have been working with a very interesting technique/technology for
mapping processes that focuses on what we call conversations for action.
These are the interactions that people have that create expectations,
commitments, or obligations. We explicitly identify who the customer is
in each interaction (be they an internal or external customer), what their
conditions of satisfaction are, and how the moves in the conversation
happen (or fail to happen!) As an analytical tool, this effectively
diagnoses the points in a process where communicative breakdowns occur.
As a design tool, it helps to more specifically define each persons role
and what they are accountable for. >From this perspective, one might
argue that the most effective measures of success is whether a customer
gets satisfied!
-- Grant Harris Workframe, Inc. (617) 497-1222 gbh@workframe.com Workframe's Home Page: http://www.tiac.net/users/gbh/index.html