GTSI, a government technology sales and integration firm reported recently that they will be missing their earnings targets and may have jeopardized their growth plans due to problems implementing a new ERP system.
GTSI went with PeopleSoft as their ERP platform (clearly prior to Oracle's acquisition of PeopleSoft). As a distribution firm, it is unclear why they went with PeopleSoft's platform for their ERP. I can see a couple of possible reasons, one being that PeopleSoft does have a fairly good presence in the public sector markets (although it's reputation there has not been pristine, with some challenged implementations in some state universities). Another possibility is that they bought when PeopleSoft was basically trying to buy business through extremely heavy discounting as they tried to fend off Oracle's takeover offer.
In any event, the challenge as expressed by some insiders and
observers was the amount of customization they needed to do to the
system to make it work for their business. This fact highlights a key
challenge with selecting an enterprise system, and really any
off-the-shelf software: picking the package that gets closest to your
business. Some people talk about an 80% "goodness of fit" number, but I
would tend to suggest that in this day and age, and with the cost and
complexity of ERP systems, a 90% fit figure might be the better one for
a rule of thumb.
When you customize enterprises systems too heavily, you really are
trying to fit a square peg into a round hole, and even if you make it
fit. the "DNA" of the system, so to speak, is still incompatible with
the company it is running. This kind of systemic incompatibility can
only cause problems over the long term. Not to impute a sense of
intentionality onto the software, but when you adapt a system designed
for one industry (say manufacturing) to another (say kitting &
distribution) you will always be fighting the software, because it has
been engineered, from the ground up, to serve this other purpose. The
costs of customization will be higher and the costs of supporting it
through upgrades will be much more significant.
In such cases, most firms would be better off purchasing a selection of the modules (say the core financials, HR, etc) and developing a customized solution to support their core business. Of course, GTSI says the reason they moved to PeopleSoft was to get off their custom system, which they felt could only be supported by their internal staff. This brings up another point: when a billion-dollar company has built itself by running on a custom system, it should increase the level of skepticism in a package meeting all that firm's needs smoothly.
Comments