IT Web reported last Thursday on an SAP Implementation failure at Johannesburg, SA. (City ‘loses’ data, Nicola Mawson ). According to the article, the city upgraded to SAP last year and integrated data from several legacy systems. Nothing unusal, except that after go live residences started experiencing drastically over inflated billing.
Moving data into SAP starts out as a simple task. Take all the data from System A and System B and put it in a single repository. After that, it often gets messy. Pressures to meet time and cost constrains often push implementers to move from fully understanding the data, and how system A relates to system B; to how to get all the data in and pass testing.
I don’t know the details of what happend on the Johannesburg project. But looking at the results, I can conclude that the data in SAP isn’t correct, there may be too much, too little, or not correctly related; and I expect the data team moved from “full understanding” to “get the data moved”, resulting in errors in SAP.
My point is that as the Project Manager, we need to be careful how hard we push the team, especially when a full understanding of data migration is necessary; otherwise, we may find the project finishes on time/budget, only to fail to meet business needs later.