Emcredit: How to transition IT from cost center to value center
Data integration and language translation bring efficiencies
A company whose product is information should regard its IT department as its “factory floor,” the place where innovation happens and customer value is produced. Yet many IT organizations in these companies struggle to make the transition from cost center to value center because they are mired in time-consuming operational processes, many of them manual, that add little perceived value. As credit bureau Emcredit discovered, however, it is possible to turn that situation around with the right software solutions and automated processes.
Emcredit is the first credit bureau in the United Arab Emirates (UAE) and enjoys official status in the Emirate of Dubai, meaning that banks are mandated to deal with Emcredit. Certified to international information security standards (ISO 27001), Emcredit obtains a wealth of data from a variety of sources: banks, financial institutions and government agencies. Like all credit bureaus around the world, its mission is to assist its clients in maintaining healthy relationships with customers while managing credit risk. It does this by providing information, reports and decision support tools that improve their clients’ evaluation of customer life cycles. However, unlike most credit bureaus, Emcredit’s scope is very broad: Emcredit provides a 360-degree view of both commercial and consumer customers.
Data: the backbone of the business
Emad Khatib, Emcredit’s CIO, realizes every day that the bureau’s reputation depends on providing accurate information on bank customers. His main responsibility is to take data from multiple operational sources and turn it into high-quality information that is easily available in the right format. “Data is the backbone of our business, so if we don’t have these processes under tight control, we will not succeed,” says Khatib. In addition, Emcredit must identify genuinely bad risks while reducing the number of consumer disputes to the absolute minimum. “You don’t want consumers and businesses challenging a decision that was made on the basis of inaccurate data, such as an incorrect date of birth,” explains Khatib. “This reflects badly on our customers, which, in turn, reflects badly on Emcredit.”
When it was in its start-up phase, Emcredit relied on manual processes to extract raw data in a specific format, which meant a lot of back-and-forth discussions with the providers about how and when the data could be delivered: after all, as commercial and government agencies, these providers had their own internal priorities; an external credit agency would have to wait its turn.
Being based in the Gulf region, Emcredit faced an additional challenge. Much of the data was in Arabic and, moreover, many of the systems generating the data ran in Arabic.
Consequently, Emcredit had to translate information into English and then back-feed it into data tables. “In fact, it was a little more complex than that. The margin of error was high because first we had to determine if the information was in proper Arabic and if it could be translated into English,” says Khatib. Only after translation could the data be cleansed, validated, standardized and merged – also manually – before it was finally stored and moved into production.
Accelerated data acquisition
Emcredit knew that its future success would require it to accelerate the data acquisition process while reducing operational costs. It could only do this by automating key processes such as the identification, extraction and translation of Arabic data and data validation. Above all, it had to minimize the effort required from the data source providers, on whose goodwill Emcredit depended.
Emcredit selected SAS to integrate financial and personal data from multiple providers and to standardize it in a single specific file format. The company considered several vendors, but SAS scored highest on the three key criteria of performance, accuracy and scalability. “SAS not only met our immediate criteria but also scored highest when we considered our long-term objectives,” explains Khatib.
“Previously we had no choice but to deal with different formats for each separate provider. Each bank would have its own specific rules for data validation and matching. SAS provided a data format that was flexible enough to accept all different source providers’ formats without requiring manual intervention. Moreover, we can add further commercial data sources without any significant effort.”
A platform for more rewarding work
SAS could also recognize Arabic tokens, which is not a straightforward process (in Arabic, different tokens can be used for words with the same meaning) and full string names (such as the full legal entity name of a company, as opposed to the commercial name). Before Emcredit translates anything through its internal translation team, it matches terms against the hundereds of thousands of records in Emcredit’s dictionary. If any tokens do not find a match, these are sent to the translation team and the new term is added to the SAS dictionary, thus progressively diminishing the amount of human intervention required.
“We have eliminated many manual processes but this does not mean we have reduced our headcount,” says Khatib. “Instead, the SAS implementation has enabled us to focus on value-adding activities.”
“Previously, we had three full-time employees permanently sitting in front of a computer screen, working on integration- and translation-related tasks. Now our employees only intervene when there is an error or an alert, and in the meantime we have retrained them to focus on customer outreach and other tasks that are more rewarding for them, and more profitable for Emcredit. This has enabled us to transition the IT department from a cost center to a value center.
“The SAS implementation has greatly increased our data coverage because we can acquire data at a much faster pace. We are also moving data much faster from staging to production, which means that the information we provide to our customers is more up to date. SAS has also improved our validation rules and processes, which has raised the quality of our data, providing a significantly higher hit rate when our customers are looking for data matches.”
Improved data quality is a huge benefit that goes to the heart of Emcredit’s business proposition: to provide banks and other clients with accurate credit information. Khatib concludes, “This enables our clients to make quick and fair decisions, minimizing disputes and helping them to build more profitable customer relationships.”
Copyright © SAS Institute Inc. All Rights Reserved.
Emad Khatib, CIO
Ensure data quality to help clients avoid mistakes when making decisions about creditworthiness of individual consumers and businesses alike
IT department transforms from cost center to value center ; clients build more profitable customer relationships with fewer disputes over credit decisions
“Now our employees only intervene when there is an error or an alert, and in the meantime we have retrained them to focus on customer outreach and other tasks that are more rewarding for them, and more profitable for Emcredit. This has enabled us to transition the IT department from a cost center to a value center.”