Absa Bank has achieved a phenomenal nine-fold return on its investment in data mining by adopting strategic analysis techniques, a clear focus and a shared IT architecture.
This gave the bank - one of SA's "big four" - a single view of its over nine million accounts operated by more than six million customers (including 380 000 who utilise Internet banking and 340 000 telephone bankers).
According to Dave Donkin, group executive: E-Business and Information Management, Absa Bank, the bank wanted to better utilise its masses of customer information to become more customer-centric and service-driven in its business practices.
"Information technology is like a plumbing system," he told delegates at SAS Institute's annual user group conference (seugi21) in Vienna, Austria, in June. "Plumbing, like information technology, is useless unless it provides clean water, in adequate amounts as and when required."
In August 2000, Absa initiated a high level information audit to assess the quality of data, levels of data duplication and redundancy of information within the organisation.
"It was clear that we had to establish an information management division within Absa if we were to unlock the value within the information and improve the organisation's performance."
Donkin said Absa's information management strategy has two clear focus areas, business intelligence and knowledge discovery. Business intelligence enables business owners to report on their own strategies going forward, and knowledge discovery unearths information gems from the bank's data stockpile.
Absa Bank uses SAS predictive analysis and Enterprise Miner to achieve this.
"If we are to use information optimally, we needed to know where to find the information," Donkin explained. "If you don't know what information you have, it can't be properly managed, sourced or analysed.
"At the same time, building awareness throughout the company of how the leverage of information supports our customer-focused strategy and educating staff on the benefits of strategic information management was vital," Donkin added. "Indeed, it's the staff's effective utilisation of the data that inevitably determined its success."
Absa's new information system now boasts over 3TB of information in the data warehouse, all of which can be used successfully to determine factors such as customer lifetime value and develop churn reduction techniques.
"We have two complementary architectures being put in place - fondly termed 'BRIAN' and 'LARA' - or Business Relevant Intelligence Architecture Nirvana (BRIAN) and Logical Analytical and Reporting Architecture (LARA) to achieve this."
The new data warehouse has also helped the bank to develop a sophisticated banking security prediction model.
"We utilised the data available to us to predict which branches have a high risk potential for robberies," said Donkin. "We then increased security at these high risk banks and have now moved from being the most targeted of the Big Four banking institutions to having the lowest cash loss rate.
"And because Absa has access to all customer information in one centralised data warehouse, we have been able to develop a focused action plan to continue improving service delivery to customers as well as streamline internal processes and systems," Donkin concluded.
Share