Business Intelligence plays a leading role in the alignment of business and IT in companies. Why? Because in dealing with data, technical know-how and content expertise are needed in equal measure. “And also, because at this interface“, as Richard Reiling,
Head of BI Development RBI IT adds, „IT must reposition itself as an advice and service facility for operating departments.“ No wonder that the evaluation of the BI enlargement at Raiffeisen Bank International (RBI) is being taken particularly seriously: A
flagship project for the whole company was intended.
||OLAP technology based on the SQL server is proving to be the deciding piece of the puzzle, integrating itself perfectly into the existing BI landscape and pushing open new "analysis doors".
| Richard Reiling
Head of BI Development at RBI IT
There was unity on the target – that was also underlined by
, Head of MIS and BI Finance and
, Head of Business Informatics. A fast and flexible Business Intelligence Solution was being sought which could expand the existing tool for standard reporting (IBM Cognos) with self-service data capabilities in the new projects (e.g.
EDWH). Expand, please note, and not replace! This is to keep the migration costs as low as possible. The subject of cost and complexity also applied to the second preliminary decision in this project. It was therefore decided, as a first step, to implement
this trend-setting pilot project only in the head office in Vienna initially.
To determine the parameters and general requirements
team had to approach the departments for whom these new capabilities were indeed designed:
„We approached the business, more precisely the finance department to, for a start find out their wishes and expectations. On the other hand it is also important in this project to consolidate access and the technology to lower
maintenance and licence costs and, as far as data sources and evaluations are concerned, to create a uniform basis.“
Even the intensive dialogue with colleagues from Controlling and Risk Management was a deciding factor for the success of the project. Only with a jointly developed roadmap in the hand was it possible to, as Richard Reiling put it,
„bring [all business interests] under one roof“. „It was of particular concern that the accompanying evaluation process would be easy to follow for everyone. Only with the support of the operating departments could we be sure
that the project would not collapse before it was even up and running.“
Now it was time to go about defining the other determining factors and milestones. Because Controlling were looking out for a planning tool, based on OLAP, and OLAP cubes were already installed in Risk Management, the OLAP technology was soon reckoned on –
just as IBM Cognos with the standardised data queries. Also, because it was wanted that the tool and content know how with Cognos, which had been built up in previous years, would carry on being used.
At the end, from the five previous leaders of the race (determined using Gartner’s Magic Quadrant), only two remained that fulfilled all basic requirements; namely, Oracle Hyperion Essbase and Microsoft OLAP.
„The scales tipped quickly into Microsoft’s favour as we began to highlight the integration and development costs more closely“
as Richard Reiling added to this position and thereby pointed out the complexity of the evaluation,
„often with BI decisions the mistake is made of basing decisions solely on functionality. This is of course important. It is, however, just as important, for example, to ensure that you have the people with the right skillset
and to get them from the market. That is by all means considerably easier with Microsoft OLAP than with Oracle Hyperion Essbase.“
In the end then, it was about much more than just functionality. How complex the evaluation was shows a supposed simple view of the cost factor.
„Migration costs, development costs, licence costs, know-how and training costs, scaling costs – Where do you want to start? All things considered,“
„the SQL server, based on OLAP cube technology has proven itself to be the deciding piece of the puzzle and fulfils all requirements: It fits perfectly into the existing BI landscape (IBM Cognos), allows a direct connection to
the planning tool (TAGETIK) and pushes open new doors of analysis for operating departments. That pays off, as well as the fact that an ad-hoc query through an Excel Front End doesn’t mean extra costs for the company.“
That it also brings flexibility showed itself in the proof of concept phase where along with the Enterprise Data Warehouse (EDWH), the network banks, which access it using push functions, also came into the project via the back door. Here the problem existed
that, at peak times (monthly reports), they kept ending up in performance bottlenecks.
„And what would make more sense“
, said Reiling,
„than to take away the necessary data for the reporting from the performance critical transaction area and likewise provide for ad-hoc evaluation over an OLAP cube? That we, at this moment, possessed the manoeuvrability to address
this for the future development critical area without bringing extra cost to the project, emphasised once again the correctness of our decision.
To avoid unwanted side effects from the new, extremely simple ad-hoc query possibilities and with it uncontrolled growth of the reports, it was made clear from the beginning that the data truth of the standardised queries remained untouched. On that point Richard
Reiling expanded, „Especially management is expecting from us that standardised definitions and queries form the basis of the reports and that they are accordingly fully comprehensible to all. Incidentally, it is also important
for the author of such ad-hoc reports to be able to align and verify their views with the results of the standard reports.“
Nonetheless, self-service data will belong to the future. Especially since, amongst the around 4000 (!) existing standard reports at RBI there are still many hidden ad-hoc reports which could be made far more cost effective (without the extra cost of the query
licence which is the custom within the Cognos world) and flexible. The aim is, in close co-operation with the operating departments, to, in the medium term, reduce the number of standard reports to 1000.
THE FUTURE / FORECAST / FUTURE PROSPECTS
Anyway, the next step on the roadmap is already signposted. There is an update on the new SQL Server 2012 in the pipeline, just like the next step in the direction of mobile reporting, to make reports accessible to management on slate devices or tablets.
Alongside this RBI is also planning to make the reports accessible to users over a BI portal. Based on which technology is still open.
„What could speak for the use of Microsoft SharePoint was definitely the Excel services,“
believes Richard, Reiling,
„which could also be made accessible online to our network banks and with it, very elegantly handle the differences in versions between the different countries.“
No matter how the decision was settled, with the scalability
of OLAP technology RBI is now on the safe side. This was clearly demonstrated in this project.
Business of IT: Interview with Richard Reiling (german) >
This case study is for informational purposes only. IN THIS SUMMARY MICROSOFT MAKES NO EXPRESS OR IMPLIED WARRANTIES.