Many companies struggle with Debbie Roth's type-2 problem -- failure to
build an operational IT system. ERP is what sofware salesmen call a
"solution", in the sense that it operates. ERP must seem like a
breakthrough to people who have struggled with the type-2 problem on an
ambitious development project.
Assuming that ERP is successfully installed (a recapitulation of the
type-2 problem, lest we forget), the organization moves on to the type 3
problem -- failure to benefit from the software. According to Landauer,
very few organizations have ever solved this problem, which is a surprise,
given the vast fortune that has been spent on IT. It isn't just a matter
of not having a good way to measure IT success: even the most forgiving
measures show no benefit. If ERP has no type-3 benefit, this would be
typical and hence no surprise (although it remains a scandal).
The ERP salesman would probably argue that ERP is an answer to several
issues that prevent the extraction of value from data, including two big
issues: data from everywhere isn't available anywhere, and data isn't
synthesized. The way the ERP salesman gets off the hook is to argue that
data availability is a necessary (?) but insufficient condition for
performance improvement, and that it is up to the organization to take the
next step. Or he could say the same thing in a finger pointing mode: we
did what we said we would do, and it is your people's fault if you can't
make it pay off.
I don't have any ERP cases to report, but I am thinking that cases are
needed because the rhetoric from salesmen and disappointed clients, and
especially the rhetoric from nanny engineers, is getting very stale. I've
grown suspicious of the following self-sealing argument: if you use
version x software engineering discipline you will succeed (never
demonstrated); people are failing; therefore people need to redouble their
efforts to use version x software engineering discipline.
I'm running a participant observer study of a $.5B project, and the
project is likely to stumble (both type 2 and 3). I'm asking: What else
promotes failure that isn't addressed by the discipline we are aspiring
to? Are people really doing poor engineering? Can any group
realistically sustain a high level of engineering discipline in the face
of political and organizational turbulence? Are the problems greater than
the means being used? Are there other ways of managing complex projects
that work sometimes but are overlooked or supressed?
It's the last question where I think there is something new to say, and
that needs case research. If anybody else is interested, I would like to
maintain a loose group of correspondents.
--Kent Myers Alexandria, VA Kent_Myers@carsoninc.com
Learning-org -- An Internet Dialog on Learning Organizations For info: <rkarash@karash.com> -or- <http://world.std.com/~lo/>