Conquering digital goods fraud with machine learning

Electronic music files, online gaming, e-tickets for events, e-gift cards: digital products are growing increasingly popular among consumers and merchants alike. For consumers, digital goods are fast, easy, and convenient. And for many online businesses, virtual products offer an alluring new revenue stream. Juniper Research cites digital goods as one of the “big three” sectors that will drive overall digital commerce to be worth USD 8 billion by 2020.

Real-time challenges

As enticing as digital goods may be, they also bring new challenges when it comes to preventing fraud and minimizing risk. One of the qualities that makes virtual products so alluring to consumers is their immediacy. Someone can conveniently purchase an electronic ticket on their way to a concert, just in time to use it at the gate. But what if that person is actually using a stolen credit card to resell the ticket for a hefty profit?

Digital goods are a ripe target for fraudsters, because they don’t require a shipping address (meaning less data is collected), and the goods are delivered immediately. At Sift Science, we typically see that digital goods receive a higher proportion of fraud attempts than physical goods.

With digital goods, the decision on whether a shopper is legitimate or fraudulent must be made in an instant. Like all online merchants, those who sell digital goods are constantly balancing the risk of approving a fraudulent transaction with risk of inadvertently inconveniencing good customers. However, unlike retail ecommerce, digital goods merchants do not have the luxury of time for manual review before shipping out merchandise. Once an order is approved, the product is gone.

Machine learning in action

To meet this challenge, many ecommerce merchants who specialize in digital goods – or who have added a digital-based revenue stream to their existing product offering – are turning to machine learning-based fraud detection solutions. Sift Science offers a Payment Fraud Prevention product that works in real time, providing highly accurate results for digital merchants.

Zentense, a Barcelona-based web and event company, uses Sift Science to stop criminals from buying event tickets with stolen credit cards and reselling the tickets for lower prices. There are two layers to the damage done by this fraud: not only do merchants face chargeback fees when the legitimate credit card holder disputes the charge, but – from a marketing perspective – reselling the tickets lowers the value of the event.

During the days before an event, Zentense processes thousands of transactions each day. In their search for a third-party fraud solution, they were looking to both reduce their intense workload and reduce the number of fraud false positives (real customers wrongly flagged as fraud).

Soon after they were integrated with Sift Science and sending real data, Zentense saw meaningful results. Within two weeks, false positives disappeared completely and the system was scoring orders bad users accurately. Perhaps most importantly, for Zentense fraud prevention was no longer a manual effort.

Want to learn more? See how Sift Science helps a restaurant booking site fight gift card fraud, and read our interview with a UK-based SVP of fraud and customer service at a video game technology company.

About Jason Tan

Jason Tan is the co-founder and CEO of Sift Science, a San Francisco-based company that helps online businesses fight fraud using large-scale, real-time machine learning. Fueled by a passion for building great products and amazing teams, Jason has also held leadership and engineering roles at BuzzLabs, Optify, and Zillow.

About Sift Science

Sift Science is a fraud detection solution for websites and mobile applications. The platform utilizes large-scale machine learning to detect fraudulent users and help online, ecommerce businesses to avoid falling victim to fraud. As the internets trust layer, Sift Sciences mission is simple yet powerful: make these online experiences faster, smoother, and safer – using the smartest technology around.