Monday, February 07, 2005
Questioning Costs of EP Project
Back in 2000 I attended a meeting chaired by Joe Salgado, then Deputy Lab Director. He made the statement that he didn’t know what ERP stood for, but it was a great idea. The Gartner Group, an IT for hire consulting firm, had convinced upper management that the total cost for a new ERP would be around $75 million and could be implemented in 4-5 years. The Gartner Group is still consulting on the ERP project, although it is now called the Enterprise Project (EP), and the delivery schedule has slipped to 7 years.
The EP Project is the Oracle based enterprise project that will replace our financial system and HR. The current costs are expected to reach $190+ million and is supposed to make the Lab more efficient and allow more dollars to flow to science.
But lets look at the history of this project. It is on its fourth project director, and the first deliverable was scaled down and released in October 2004. The project continues to underdeliver while project costs rise due to the large number of outside consultants working on this. Savings were projected at around $20 million per year, and if those actual savings can in fact be realized, it will be 10 years before the project breaks even.
What was wrong with the old systems? We got paid, we were able to purchase items and have them delivered quickly. It is the palace building of our senior management that feels a need to use taxpayer monies to construct shrines similar to what we recently saw in
The biggest problem with ERP implementations is politics. A whole lot of people have serious investments in the status quo. Many others resist simply because they know it won't be the same as what they're used to. Finally, management can shoot itself in the foot by changing the scope in midstream. All of these occurred.
For example, when Nanos came in, he trashed the incremental approach recommended by every qualified professional around in favor of a "big bang" implementation because he'd been "successful" with a big bang approach in the Navy. Or so he thought. Turns out people were blowing smoke up the Vice-Admiral's skirt about that particular project. Cha-ching! Time to re-scope! Consultants love this stuff.
After a couple of project directors went bye-bye, Nanos finally saw the light and relaxed the big bang requirement. So, the project was back to the incremental approach, and had to restructure once again. Cha-ching!
Again, it was management that forced the first release to go out with inadequate testing last October, just to stay on a political schedule. Cha-ching! Testing and fixing during production are much more expensive!
Don't blame the project. LIke any other software effort, ERP implementations make managers pay through the nose for every silly political act they impose on the effort.