Improving Performance and Scale for Critical Risk-Scoring and Anti-Fraud Processes

Pagantis provides automated, consumer finance for ecommerce transactions. Founded in 2011, Pagantis has grown steadily in response to relentless demand for faster and more flexible payment methods.The Pagantis point-of-sale platform allows consumers to pay for goods and services in monthly installments with a fully automated, paperless process and provides ecommerce merchants with a simple onboarding process to offer consumer credit in conjunction with e-commerce purchases.The company has provided over €500 million in online consumer loans on its innovative platform, which has been developed by leveraging the company’s expertise in data science, technology, regulatory compliance, and finance.

The Challenge

For Pagantis, the speed of onboarding each new customer is essential as this is how the company maintains its competitive differentiation. At the heart of the Pagantis platform is an algorithm that analyses the risk of fraud and credit, using big data and machine learning techniques, to approve or disapprove new loan requests as fast as possible. Pagantis recognized that it needed to make improvements over its existing relational database structure in order to deliver the near real-time approvals process expected by clients and consumers

The company examined a number of alternatives in the hope of finding one that offered a client-focused approach, state-of-art technology and next generation database management solution that integrated seamlessly with its existing workflow.

The Solution

Pagantis is using TigerGraph to calculate a customer’s credit rating across their real-time activities as well and all available historical data. A graph database is the only data model where each customer’s data entities are pre-connected offering a simplified way to analyze complex relationships. Performance is enhanced by TigerGraph’s native parallel graph, which focuses on both storage and computation, supporting real-time graph updates and offering built-in parallel computation.

Pagantis deployed TigerGraph on AWS and the result is a scalable, high-performance system that allows Pagantis to quickly deliver real-time insights into complex relationship-based workflows that are common in tasks such as credit scoring, fraud detection, recommendation engines and risk analysis.

The Results

With TigerGraph, Pagantis has been able to improve and accelerate its anti-fraud and risk process and consequently reduce users’ wait times significantly. By offering a faster customer experience the company is attracting more new business than ever before with both third-party clients and end customers.

“We examined a number of alternatives but only TigerGraph offered a client-focused approach, state-of-art technology and next generation database management solution that integrated seamlessly with our existing workflow,” says Martynas Sukys, Product Owner at Pagantis.

By switching to TigerGraph deployed on AWS, Pagantis was able to significantly reduce the delays associated with critical processes such as fraud detection and risk scoring and create a foundation to deliver its service at scale.