With sophisticated fraudsters conducting repetitive, small-deposit account transaction fraud attempts, it is increasingly difficult for financial institutions to detect and prevent fraudulent activity. To successfully thwart these attempts, it is important to start prevention at the point of disbursement and with automated verification and fraud detection at all points in the clearing process. Newer generation technology can detect fraud more efficiently, resulting in improved detection accuracy and optimized labor utilization.
A National Top 5 Bank turned to ARGO’s fraud solution, OASIS™ (Optimized Assessment of Suspicious Items), designed to provide cross-channel, multi-fund analytics and adjudication workflow to detect fraudulent transactions and suspicious items. Using OASIS advanced machine learning models, the institution was looking to:
The bank was reviewing an average of one to two percent of their check volume as daily fraud suspects, and had significant losses caused by the inability of their legacy system to detect counterfeits from metadata analysis alone. The suspect-to-fraud ratio for their rules-based legacy system was 250:1.
Using machine-learning pattern-detection models, OASIS performed image analysis (signature and check stock verification) using reference images for comparison. The OASIS solution demonstrated the ability to provide significant lift in the detection rate of counterfeits and forgeries/forged maker signatures with detection rates greater than 90 percent. Additionally, with a suspect-to-fraud ratio improved to 60:1, OASIS significantly reduced labor needs and costs.
For more information, download the “Industry Proven Results” interview with David Engebos, President and COO of ARGO.