Voice of the Industry

How can the fashion ecommerce sector fight fraud and grow its business at the same time?

Friday 12 June 2020 09:16 CET | Editor: Stefana Ivan | Voice of the industry

Nethone explains how Machine Learning and Data Science can help the fashion ecommerce sector to fight fraud and increase business 

A fashion ecommerce platform needed to secure its online global traffic, but that was just one part, as the additional challenge was to increase conversion rates, while lowering both the overall chargeback rate and the amount of manually-reviewed traffic. Check out how Nethone managed to meet all of these goals with Machine Learning and its proprietary Profiler technology, in the following article.

The online fashion and luxury goods ecommerce space attracts a great deal of credit card fraud and chargebacks. Carefully tuned Machine Learning models can drastically reduce the occurrence of both. 

The problem 

The fashion ecommerce platform faced 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. 

After consultations with the client and setting up a detailed plan of work, Nethone managed to fill all of the above KPIs during 4 months of full integration. 

The solution 

Nethone is a Know Your Users (KYU) company that uses Machine Learning technology to detect and prevent card-not-present fraud, including protection against account takeover. Moreover, the Nethone proprietary Profiler is able to enrich the context of the user’s understanding with over 5000 attributes, including device fingerprinting. All of them are processed into recommendations in real time, without affecting the UX of your service. 

The fraud prevention solution for this ecommerce platform was based on the above custom technologies, and was implemented as the following: 

Nethone takes an iterative approach to model building. New models are built that target the incorrect predictions of previous model builds. 

1. Machine Learning models 

To detect chargebacks, the Nethone Data Science team implemented two supervised Machine Learning models. The goal of the first one was to detect chargebacks directly, while the second one was to replicate the results of the decision process of previous anti-fraud providers. 

2. Tests and XGBoost models 

Then, the DS team trained and tested multiple Machine Learning models using various statistical techniques. Among them, two acutely-tuned XGBoost fraud detection models achieved the highest performance, so they were quickly implemented.  

Both models’ performances were monitored in production. Moreover, they were adjusted on a monthly basis to match the fashion company’s internal KPIs. The team scaled its data processing resources to ensure both models’ predictions would not experience any downtime during important sales seasons, such as Black Friday or the winter holiday season. 

3. Optimisation and fraud detection models 

In the next stage, Nethone deployed minor optimisations to its in-production models. Due to these updates, fraud detection capabilities for risky geographies were expanded. In addition, it also targeted previous models’ false positive rate to ensure that it only rejects fraudulent transactions. 

The result 

Thanks to the above-described actions, Nethone cut down the platform’s chargeback and manual review rates. 

The manual review rate was lowered by nearly 60%. At the same time, the percentage of accepted traffic increased in comparison to the rates of other anti-fraud solutions providers. 

The overall chargeback rate decreased by over 12%. For a fashion ecommerce platform’s most valuable brand, the chargeback rate was lowered by as much as 89%. 

The company’s internal KPIs were met – it kept its overall chargeback rate below 0.5% for all of its brands. 

Finally, as Nethone became the primary fraud recommendation system for the client, it significantly cut additional costs and removed the burden from its fraud management team, which resulted from the growing amount of manual review traffic generated by other anti-fraud solutions providers in the past. 

Reduction in manual review and chargebacks means lower costs for the company. Transactions with reduced friction improve the customer experience, which translates into customer loyalty and return purchases. 

About Nethone 

Nethone is a global provider of AI-driven KYU (Know Your Users) solutions company that allows online merchants to understand their end-users and prevent them from committing online fraud. By using ML technology, Nethone is able to detect and prevent card-not-present fraud, including protection against account takeover. Founded in 2016 by data scientists, security experts, and business executives, Nethone successfully cooperates with ecommerce, digital goods, travel, and financial industries on a global scale.


Free Headlines in your E-mail

Every day we send out a free e-mail with the most important headlines of the last 24 hours.

Subscribe now

Keywords: Nethone, fashion ecommerce platform, KYU, Know Your Users, ecommerce, Machine Learning, ML, data science, Profiler, UX, XGBoost fraud detection models, chargebacks, KPIs
Categories: Payments & Commerce | Digital Identity, Security & Online Fraud
Countries: Europe
This article is part of category

Payments & Commerce