Case Study: Top U.S. Bank achieves real-time fraud detection with SingleStore

Top U.S. Bank - Customer Case Study

Top U.S. Bank needed to stop rising credit card fraud and replace a slow overnight batch process that let fraudulent transactions slip through during the day. The bank also wanted a more flexible, in-house fraud detection platform that could update models quickly, support standard SQL, and deliver a better customer experience. SingleStore was selected as the data platform to support this real-time fraud initiative.

Using SingleStore with Kafka-enabled streaming ingest and SQL-based analytics, the bank built a system that runs 70+ queries in parallel in under 50 milliseconds and makes approve/decline decisions within one second of a card swipe. The result was faster fraud detection, improved customer experience, and the potential to save tens to hundreds of millions of dollars in fraudulent transactions.