Thursday, October 26, 2006

What does the P in CPM really stand for?

At the start of any CPM initiative, the future tends to look bright for everybody involved. There is great vision and leadership, major benefits to be achieved, and the best possible software to support the initiative.

After a few months however, the initial enthusiasm usually begins to decrease. Senior executives have new issues to worry about, project team members are swamped with everyday routines, and the software appears to be unable to fulfil all requirements. The project that was so well managed in the beginning is now merely muddling through. New requirements are creeping into scope, the software is being tweaked to the max and beyond, and the original strong vision seems to be long forgotten.

But a few project trips, some executive pep talk, several stiff conversations, and a lot of hard work later, the project is finally ready to 'go live'. All is well that ends well! Or is it? Finance staff now seems to be having a lot of issues getting their data into the new system. Remote locations can't even connect. Producing reports is taking longer then ever before. The system keeps crashing for no apparent reason. It is slowly becoming clear what the P in CPM really stands for...managing the performance of the CPM system.

The first step is usually to perform some stress and performance testing, remove a few obvious bottlenecks and simpy upgrade the hardware. Most of the times this is not enough. As the systems grows over time, new performance issues arise that prove even more difficult to resolve. Improving performance is a war that has to be fought on several frontlines at the same time. More than a state-of-the-art IT infrastructure, it involves hard decisions on the architecture of the whole CPM system, trading off functionality with performance, changes to the process and organization underlying the CPM system, and, yes, some more stiff conversations.

In hindsight it is always easy to explain the poor performance. It usually comes down to taking more time to think things through at the start. Are all the must-have requirements really that critical? What else do we want from the system after the initial implementation? What is the simplest way to start? Why don’t we say goodbye to some of the dysfunctional practices that have developed over the years? These are not easy questions to answer. But answering them is critical to the ultimate success of the CPM initiative.

To end this blog on a positive note: there are a lot of experienced people out there that can help you in answering these questions, and making your CPM future look brighter.

Author: Rutger van den Berg (