Over the past few months I've come across an increasing number of organisations currently using SAP BusinessObjects Planning and Consolidation version for Microsoft (SAP BPC for MS) who are contemplating migrating across to the SAP NetWeaver version of the product. Whilst there are certainly benefits to doing so, which I'll go into in this blog, it is not a simple process, and some organisations I've discussed this with have ultimately decided that the benefit does not justify the cost.
What types of organisations are looking to migrate and why?
The typical organisation looking to migrate will...
have a mature SAP BPC for MS implementation on an old version, perhaps even one of the legacy Outlooksoft versions
be using SAP ERP as a general ledger and will be facing integration challenges in moving data from SAP ERP to SAP BPC
have requirements to extend the functionality of their existing solution and as part of that process are re-evaluating their systems architecture.
In evaluating the SAP NetWeaver version of SAP BPC, these
organisations perceive the following benefits over and above the
Better integration with SAP ERP - Being based on SAP NetWeaver means that BW's sophisticated ETL capabilities and out-of-the-box integration with SAP ERP can be leveraged to streamline the integration of data (transactional and master data) from SAP ERP to SAP BPC
Maturity/ functionality - SAP BPC 10 for NetWeaver represents a great leap in terms of both functionality and maturity. Moreover, whilst SAP has consistently promised to develop the Microsoft version in line with the SAP NetWeaver version (and has been good to its word), there are some cases of functionality being available in the SAP NetWeaver version that is not in the Microsoft version. There is also a common perception amongst traditional SAP ERP clients (who happen to have the SAP BPC for MS version) that the sap NetWeaver version is naturally where the product fits for a large enterprise and that it is likely it will be the version that gets the most attention from SAP
SAP ABAP - The SAP NetWeaver platform enables the development of planning functionality using SAP's ABAP programming language. This is more flexible and better performing than the equivalent on the Microsoft version and enables the delivery of more sophisticated, better performing and generally beneficial planning models
SAP HANA - Organisations looking to migrate generally have well-established budgeting and forecasting solutions, but are looking to move along the maturity curve and enhance their solutions to enable them to perform complex scenario planning, driver-based planning, introduce more automation, and so on. What has previously held back the implementation of such solutions was the ability of the underlying hardware/ database/ software to deal with the increased data volume and volume/complexity of real-time calculations required to make them effective. SAP HANA means that providing such solutions with acceptable system performance is now possible. SAP HANA is only available on the SAP NetWeaver platform (and I am unaware of any plans to make it compatible with the Microsoft platform). This is therefore driving interest from existing SAP BPC for MS customers to SAP BPC for NetWeaver
Reduced TCO - For organisations already using any of the other NetWeaver based SAP tools (e.g. SAP ERP, SAP CRM, SAP SRM, SAP BW) SAP BPC for NetWeaver represents an opportunity to consolidate onto a single platform, with all the benefits in terms of reduced maintenance and support costs.
Technical migration or redesign?
The first dilemma to grapple with is whether the migration is purely technical from the MS platform to SAP NetWeaver, or whether a redesign is desirable. Certainly, there are many strong arguments to redesign at least portions of the solution:
Not redesigning the integration with source systems to make full use of BW's ETL capabilities would represent a missed opportunity to drive process benefits
Not all of the Script Logic/ MDX syntax available in the MS version is supported in the SAP NetWeaver version. In any case, with ABAP there is the possibility to improve performance and functionality of calculations making it possible, for example, to make some calculations real-time that were previously performed in batch due to performance considerations
Do the enhancements envisaged point towards an amendment of the data model? For example, can planning be performed at a more granular level (e.g. SKU instead of product group) now that SAP HANA removes some of the obstacles to having a large planning data set?
Of course a redesign is more complex and costly than just a technical migration, but the benefits of moving to SAP NetWeaver can only be fully realised if at least a portion of the solution is redesigned. Having worked through this with several organisations, my conclusion is that there is no business justification for a simple technical migration.
Of course, even a redesign does not mean starting entirely from scratch and it would be beneficial to start from a technical migration and modify, rather than build from the ground up. However, due to there being no migration tool to bring across the SAP BPC for MS objects (data model, dimensions, reports, input schedules, script logic, security model, BPFs etc.) you are effectively forced to regenerate the SAP BPC environment manually (or copy and paste where possible).
As an aside, there was (some years ago) a migration tool that would migrate a MS AppSet to the NW platform. This was, I believe, developed by SAP Services and, in order to get access to the tool, you had to attend a training course in how to use the tool. Recently, I approached SAP about this tool, but it appears that it is no longer supported (it was developed for SAP BPC 7.0/ 7.5 and was never rewritten for SAP BPC 10) and the training courses are no longer offered.
The greatest effort usually comes when migrating across script logic, especially in a complex environment where there is a great deal of it. Assuming that you are to take the approach of converting all script logic to ABAP, then the main challenge is to find resources who understand both SAP BPC and your planning model and can code ABAP.
Significant effort is also required for effective regression testing and deployment of the solution, without impacting the business processes that the legacy application was already supporting.
In short, the effort required to perform the migration should not be underestimated. I recently estimated out on such migration (the application was fairly complex with multiple applications, 100 reports and around 90 script logics) to be in the order to 700 man days effort.
There must be a solid business justification to perform a migration from SAP BPC for MS to SAP BPC for NetWeaver as this is a costly and complex task. Simple 'technical migrations' rarely, in my experience, give the benefits needed to justify the costs. Organisations looking to perform this migration should look to the enhancements to their current planning processes that they could only achieve through such a migration and the benefits they provide. Certainly, SAP HANA means that long understood best practices are now becoming feasible to implement and can deliver true competitive advantage.