How a leading card issuer used LogicLoop for fraud operations

One of LogicLoop's customers is a prominent card issuer, aka they help developers create virtual and physical payment cards, and monitor their transactions. They raised a new round of funding recently, which fueled surging demand for their product. They surpassed $1 billion in annual processing volume and 4x growth on its API developer platform in just a few months post-launch.

‍

Before LogicLoop

The card issuer's compliance and fraud team was using an industry-leading rules engine vendor to manage onboarding and ongoing monitoring. However, as their business grew, they needed more: 

• They couldn’t export transaction data out of the platform, so had to repeatedly ask for Google Sheet exports 
• They could not experiment on rules easily, or run analyses outside of what their rules management interface could support. It took them 6 months to add a rule that looks at spending > 3 stdevs away from normal, something that should be table stakes. Besides, their best risk analysts wanted to work with data directly, often in SQL, and did not want to learn a special purpose language with new jargon. 
• They did not have a flexible case management system that let a risk or dispute operations member review and escalate cases quickly. 

‍

Their Risk Operations Lead was in risk monitoring at Stripe earlier, where he used state-of-the-art internal tooling and would single-handedly thwart millions of dollars of potential fraud in a single quarter. As a fast-growing startup, the card issuer did not have the technical resources to build a good risk alerting system in-house. They had to be very careful to protect good users and stop bad users, especially around launching new products. Around one such product launch, fraudsters launched a huge attack when some of the risk team was taking a (much needed) vacation. He wasn’t able to alert the eng team to update fraud rules, causing unprecedented fraud losses in the several hundreds of thousands of dollars. They knew something had to change. 

‍

Using LogicLoop

The card issuer was most interested in having a flexible system to: 

• Detect suspicious activity in minutes, not hours or weeks 
• Experiment on new rules without waiting for vendors or engineers
• Manage suspicious activity review and dispute cases efficiently

‍

Specifically, they continue to use their original fraud engine vendor's integrations for some compliance and KYC workflows, but started using LogicLoop for user and transaction monitoring. Some example workflows: 

• Large, declined or refunded transactions
• Block high risk country, MCC and OFAC transactions
• Ensure compliance with velocity and other rules 

‍

See a list of fraud and risk monitoring templates here.

‍

The card issuer found that LogicLoop was easy to integrate and onboard (in weeks not months), gave their analysts more control than any existing vendor without having to build themselves and very responsive. 

Using LogicLoop, they were able to reduce review time by 75% and hence headcount, detect fraud within minutes that reduced losses per attack 5x and increase the pace of product delivery. 

‍