Overview
Our client is one of the leading global platforms for the luxury fashion industry. It connects customers in over 190 countries with over 1,200 of the world’s best brands, boutiques, and department stores. The company reached out to Nethone with three issues, which had to be addressed on a global scale:
- the continuous growth of manual review of traffic, which was generated by legacy anti-fraud solutions providers,
- high chargeback ratios - the company set a bold KPI to keep them below 0,5% for all of its brands,
- weak chargeback risk detection, which translated into growing financial losses.
Thanks to our Machine Learning models backed by our proprietary Profiler, Nethone decreased its chargeback and manual review rates for the client’s key markets in just under 4 months of full integration for their US, China, and UK customer bases.
Challenges
Solution
The fraud prevention solution Nethone developed for the client is based on custom technologies, including in-depth user profiling, device fingerprinting, behavioural biometrics, and advanced Machine Learning. Nethone helped to slash chargebacks and strengthen the rapid growth of the company while ensuring a seamless user experience for desktop and mobile customers. The accuracy of all recommendations is established by Nethone’s proprietary Profiler, which offers unmatched depth and breadth of online user profiling capabilities (5000+ data points, mostly non-declarative, proprietary data extraction methods regarding hardware, software, network, behavioral contexts).
Nethone's Data Science team implemented two supervised Machine Learning models to detect chargebacks. One was built to detect the risk of chargebacks directly, and the second was created with the intention of replicating the results of the previous fraud prevention solution's decision-making process. The DS team found the errors of the previous model and iterated from there. As a result, we managed to slash the manual review rate by nearly 60% while increasing the percentage of accepted traffic in comparison to competitors’ rates of accepted traffic.
The Nethone Data Science team also developed several models built on sensitive segments of traffic that were the most important for the client in order to detect fraudulent interactions and prevent them from becoming chargebacks. The overall chargeback rate decreased by over 12%. For the fashion eCommerce platform’s most valuable brand, the chargeback rate was lowered by as much as 89%.