Data quality is major issue for financial institutions

The costs of inaccurate data can be enormous - and not only in monetary terms. The responsibility for squeaky-clean data must therefore lie with everyone within the organisation, from the receptionist to the chairman.

For South African financial services companies, data quality is a particularly major issue. In light of increasing fraud threats and new regulatory requirements, banks are placing major focus on the quality of their information.

"We see information quality as a cross-group process. Clean data is dependent on the contribution of all the role players," says Cornie Victor, general manager of the Business Enablement department in Absa's Information Management (IM) Division.

Recognising that the consequences of bad data quality can be catastrophic, Absa's data quality initiatives, which are underpinned by SAS business intelligence technology, started in 1997.

Inaccurate or incorrect data can call into question the production of performance indicators, reduce the credibility of the information system and even lead to significant financial losses.

"In fact, dirty data can damage every aspect of a business," says Annemarie Cronje, Solutions Architect at SAS Institute SA. "On the customer-side, outdated information or an incorrect credit score could mean a failed marketing campaign or an angry customer. In the supply chain, poor product data can cause production bottlenecks and slow down delivery orders to retailers."

Accurate, up-to-date information on customers not only improves customer service, but reduces the risk of an expensive marketing campaign failing - not to mention opening up new opportunities to cross and up-sell.

Victor says that if incorrect address information led to only 2% of Absa's annual customer mailings being returned, this would cost the bank close to R3 million each year.

"One of the major benefits of excellent information quality is a higher response rate to marketing campaigns," says Victor.

"Others include improved customer loyalty and service, improved shareholder value and opportunities for new product development."

Victor believes that data quality is a journey: "Organisations need to create awareness of quality on an ongoing basis, measure constantly and benchmark against latest trends."

His department, Business Enablement, does not take ownership of the bank's data. Instead, he sees it as the custodian of information quality across the information process.

"The department also monitors the quality of data captured in our systems, co-ordinates the group-wide data clean-up initiatives, and creates overall awareness for information quality," he says.

"Our initial forays into benchmarking our capability indicate we are well positioned from a data quality process perspective. We are approaching level four on a five-point scale using a Meta Group benchmark."

The Meta Group's Information Management Maturity Scale covers a wide range of data quality issues including data ownership, policies, processes, measurement and technology.

Victor's advice for companies embarking on a data quality journey:

* Ensure training material is adequate and staff trained accordingly.
* Ensure processes and procedures are documented and implemented.
* Ensure auditing of processes, training, policies and procedures.
* Ensure system validations are implemented.
* Ensure proper governance structure is in place.

"In selecting a data quality technology solution, companies need to look for sophisticated matching and standardisation capabilities that enable users to analyse, clean and standardise data across all platforms," concludes Cronje.