Fighting fraud in near real time
Hungary’s OTP Bank gains competitive advantage via data visualization, fraud detection and social network analysis from SAS®
When a Hungarian banking group began experiencing an uptick in fraudulent activity, it knew it needed a better way of spotting potential fraud before the credit was issued. “More sophisticated criminal techniques are making fraud detection and prevention much harder,” says Zoltan Zsolt Nagy, Head of the Analysis and Modeling Department, Credit Approval and Risk Management Division of OTP Bank. “We needed a solution that could return quick analysis on customer behavior reliably across multiple accounts and systems.”
The portfolio analysis team is now able to generate reports in 10-20 seconds; previously, this activity required an entire day.
Zoltan Zsolt Nagy
Head of the Analysis and Modeling Department, Credit Approval and Risk Management Division
OTP Bank is the biggest independent banking group in Central Eastern Europe. It offers banking services to more than 13 million customers in nine countries. Fraud was leading to direct losses and would have eventually forced the bank into a less competitive position, as it would need to charge higher rates to offset losses.
“We already used fraud prevention techniques, such as application, behavioral and transactional scorecards, but we were looking for something more effective and innovative, such as a system that would allow the analysis of social networks,” Nagy says. OTP Bank chose a solution consisting of SAS® Visual Analytics and SAS® Fraud Network Analysis, which includes social network analysis.
Beyond the traditional prevention
The solution uses a hybrid analytical approach to fraud detection, helping reduce false positives by incorporating the latest anti-fraud measures, advanced statistical models, and deterministic and associative analysis. The social network analysis component examines links and associations among counterparties and the various elements from the individual cases reported, highlighting potentially fraudulent practices. When a fraud referral or case is completed, the results are stored to use in future models, which improves performance and efficiency.
Speed of analysis and visualization
In the past, attempting to dig through the data was a time-consuming process. “The speed of analysis is an essential factor for our business,” Nagy says. “The system must be able to immediately provide us with results on the customer situation, enabling us to make faster and more reliable decisions on the loans we grant. Extremely sophisticated solutions are useless if the results take too long to be processed,” he adds.
The visual element is critical because the data is used throughout the bank by staff who need to explore the data from multiple angles – not just look at static reports or charts. “We particularly appreciate the ability to integrate with the OTP loan origination system, and we value the visualization component, which enables deeper detailed analysis, even for non-technical staff,” Nagy says. In addition to the bank’s IT team, the SAS solution is also used by security staff in charge of fraud detection, by investigators who authorize credit and by risk managers who monitor reports.
“For analysts, it is easy to explore and process data, even by simply using Microsoft tools; however, decision makers and risk managers need more intuitive, user-friendly tools so they can obtain the information they need quickly and accurately,” explains Nagy. “Thanks to SAS Visual Analytics, the portfolio analysis team is now able to generate reports in 10-20 seconds; previously, this activity required an entire day.”