NatWest and Featurespace worked together to pioneer a bespoke transaction monitoring framework for payment fraud attacks.

Challenge

NatWest first engaged Featurespace in 2019, when its incumbent fraud detection system was struggling to identify fraud and scams effectively. Wanting a more efficient way to identify existing and new types of fraud, the bank sought a real-time fraud detection platform that incorporated adaptive machine learning models to keep pace with fraudsters' evolving tactics.

Solution

The bank chose to upgrade to a real-time fraud detection platform, with adaptive machine learning models as their first line of defense. The bank also required the platform to offer the flexibility to design, develop and deploy fraud risk modeling based on advanced applied analytics.

Results

NatWest’s deployment of Featurespace’s solution achieved results from day one. After an enterprise-wide deployment, the Featurespace Platform surpassed expectations with its ability to detect and reduce third-party fraud losses and overall fraud. NatWest saw an increase in the value of fraud and scams detected including an immediate decrease in false positive rates (genuine customer activity declined), within just 24 hours of deploying the Featurespace Platform.

In addition to meeting other objectives, including successful integration with core payment systems and response times with established performance tolerances, NatWest’s new platform achieved marked improvement in detection rates over incumbent solutions.

2020 saw NatWest Group release their 2019 Results & Investor Update, listing Featurespace as a “strong partner” as part of its ongoing strategies. The results of the enterprise-wide Featurespace platform deployment has recently elevated NatWest as winners of the “Best Security and Anti-Fraud Development” at the Card & Payments Awards 2021.

Success with Automated Deep Behavioral Networks (ADBN) 

Building on the significant improvements in APP scam detection, NatWest turned to Featurespace to help enhance real-time debit card transaction fraud detection by replacing its previous fraud risk score and transaction monitoring platform.

The bank has incorporated Adaptive Behavioral Analytics, ADBN and ensembled risk scores into its overall rule set for handling debit card transactions. The solution integrates directly with the bank’s customer communication platform that sends SMS alerts to customers to approve or decline transactions.

NatWest expects to see further improvements over time through the application of the ABA, ADBN and ensembled risk scores.

The Featurespace Platform enables NatWest to:

• Improve the fraud account detection rate (ADR) by +3% (so fewer customers experience fraud losses).
• Detect cards being added to file and wallet enrollments for the first time— reducing the number of customers experiencing fraud losses to around 3,000 per month.
• Reduce false positives from 11:1 to 8:1 on average.
• Catch fraud at an earlier stage, reducing the number of fraudulent transactions per case from three to two.
• See an improved Net Promoter Score (NPS) of above 70% positive.¹