Below is a short extract from the article. Please log in to download the entire article or to comment.

.....The first common blooper around SAPBI implementation is that it gets implemented (in parallel) with ECC (SAP ERP’s latest version) – or for that matter with the source system implementation. Principally, SAPBI gets data from its source system, it stores the data and uses it for analytics. If the source system itself is not stable (i.e. it is going thru its own implementation), it will not be able to provide consistent data to the (under implementation) SAPBI system – so logical! Or, think about it, if the datasources in the source system that are being extracted in SAPBI system are changing in the source system, their definitions (and associated data targets) in SAPBI will have to go thru changes each time. Which means, required configuration changes in the source system (however minor they are) will lead to frequent changes in SAPBI system – does that make sense? Isn’t it waste of time and effort and money? Best approach – which is not always possible – is to implement SAPBI AFTER source system implementation is over and AFTER source system is stable. Now, since this is not possible always, the least one could do is, let the source system implementation make progress. For example, let the source system configuration at least enter user acceptance testing before starting development of SAPBI. This way, SAPBI will catch up with the source system implementation.....