Last week, I was interviewing one of the sales executives in banking for one of our newsletter items.
One of my favorite questions is to ask - 'What according to you is the most innovative solution in SAP FS?' He talked a lot about how HANA and Mobility has brought in a lot of innovation and how it helps to improve customer centricity etc... I was a little surprised and asMy immediate question to him was that why didn't he mention bank analyzer or FSN ? I have even heard in some of my previous interviews, people mentioning POWL as a feature developed in SAP FS.
His reply was..FSN is gaining momentum but Bank Analyzer!! Oh, it's a black box.
Bank Analyzer - A black box?
This became my lunch time discussion with a BA team architect and IFRA was the main discussion topic. As we all know Bank Analyzer is based on Integrated Finance and Risk Architecture (IFRA). IFRA is modular, service-oriented and has a layered model and we started listing down all the nice features of BA.
In SDL, preconfigured template products are available for several bank products. Defining a new template version or enabling a new attribute in the template is an easy process using the service module architecture. Source data services provide a variety of features to read data from primary objects in SDL. One
doesn't need to know how or in which tables the data gets stored. Earlier in my blog I discussed about Source data services in BA.
http://scn.sap.com/community/banking/blog/2013/10/20/source-data-services-in-bank-analyzer
PML Layer covers the processes like calculation of credit exposure, hedge adjustments, valuation of business transactions, etc. and a method such as calculation of fair value / cash flows. GAAP specific business content covers most of the necessary posting schemas. Even if one need to create any additional postings, the linking between calculation step to item type to posting key figures is clearly documented and can be easily be implemented. Risk calculations are implemented using the module editor architecture.Any additional logic can be implemented by creating a new version.
RDL has a hierarchical structure consisting of result types and each of them representing the semantic unit. Also , there are several predefined result
categories and result node categories to give the possible options for customizing the RDL.
The predefined business content covers a lot of complex processes across all the layers.
And we concluded our lunch saying - Bank analyzer is not a black box.It's quite transparent. And when went back to our desks one of the colleagues was waiting who had to solve a rounding issue in BA as amounts were not allowed to be posted to a G/L account in decimals. All he wanted to know was where to put the math function for rounding-off.
We just can't do it. We cannot alter the amounts of the business transactions in PEBT. It has to be done in USBT. So the solution we proposed was: create a posting key figure for these round off amount. create a new calculation procedure , calculation template, calculation step, calculation method. Find the relevatn item template. Link all these settings and post the amount to an intermediate account. Write the round off logic in the calculation method ; Post the difference and finally do a move to the posting key figure linked to the final account.
May be it's too overwhelming for a round-off issue but i couldn't think of any easy way to acomplish it.I am forced to think in a structured way as BA is structured.So.. I started to reconsider my earlier opinion about BA. Though BA is transparent, it is an over-engineered product. It's not a black box but it’s a cramped box with these huge customization.
But BA not only confines itself as a sub ledger but caters to many functionalities like Basel , credit risk , ALM etc. And when it tries to address complex functionalities , it tends to become an equally complex structured software.
What do you think ?
