Case Study: Leading US-Based Card Issuer achieves 40% reduction in false positives and 50% increase in fraud detection with DataVisor

Leading US-based Card Issuer Reduces False Positives by 40% and Increases Detection

Leading US-Based Card Issuer, a $1.4B consumer banking division rolling out online and mobile checking, savings, ACH and wire services, faced a legacy rules-only fraud stack that missed fast-evolving fraud rings, generated high false positives, and forced analysts to spend 10+ minutes per case. The issuer selected DataVisor and implemented its ML-powered fraud detection and case management in a two-phase deployment to augment existing rules and protect the new digital channel.

DataVisor deployed a combined Unsupervised and Supervised ML approach with an upgraded Rules Engine, Identity Graph and a customized Case Management system (data stitching, enhanced intelligence and configurable review templates), streamlining reviews and automating orchestration. DataVisor’s solution reduced false positives by 40%, increased fraud capture by 50% without adding reviewers, cut review time to under 1 minute (accelerating decisioning by ~72%), and delivered >95% review accuracy.