While most of the attention in the media goes towards ERP project overruns (which friend Mike Krigsman faithfully catalogs) the reality is the bulk of ERP TCO comes after "go-live".
The 3 big components "post-live" are annual vendor maintenance, upgrades and application support. Customers are increasingly focused on squeezing maintenance as this blog catalogs on a frequent basis. Customers have also learned to slow down the rollout of major upgrades - be they to Microsoft Vista or SAP ECC 6.0.
But the third component - application support - continues to be stubbornly high. And it has not seen much analyst or media scrutiny. So I was pleased to see Frank Scavo is about to present his firm's findings on the topic at an APICS seminar this week. Having seen an advance copy of his slides, he will be presenting a chock full of stats and some non-intuitive findings. Probably the most important stat - on average each ERP support staff supports roughly 40 named users. Do the math on that labor cost over a "post-live" 8 to 9 years - and you can see what I mean by "long tail" of ERP TCO.
One way to manage that cost in the last few years had been to use offshore firms. In the run-up to Y2K they showed themselves adept at transitioning knowledge on a number of relatively steady, legacy custom developed apps. They have since extended that capability to ERP support, particularly around SAP and Oracle applications. But with the currency and wage inflation most Indian firms have experienced in the last couple of years, that support cost is re-emerging as an area that deserves significant CIO attention - and, of course analyst and blogger scrutiny.
I have invited Frank to contribute a "Real Deal" guest column to share more of the findings of his firm's study. I will also share in future automation and productivity tools that have been used to manage that "tail" and where the industry could improve significantly.