How Did PetSmart Eliminate $12M in Fraudulent Transactions?

10-July-2018

PetSmart, like many online retailers, frequently suffers from fraudulent transactions on its website when stolen credit card numbers are used to purchase products. But the company is getting smarter. Last year it stopped about $12 million fraudulent transactions from being shipped and this year the company is on pace to prevent even more. How?

The answer is an important consideration for anyone doing business online.

According to this report on ZDNet, PetSmart, and many other large retailers like Staples, Dunkin' Donuts, Domino's Pizza and Crate & Barrel, are implementing new technologies from software companies that use machine learning and artificial intelligence. In PetSmart's case, the technology they use is provided by a software company called Kount. The software is used to compare a single transaction to millions like it and then look for signs that the transaction may not be on the up and up.

Kount's technology - called BOOST - aggregates millions of transactions and their outcomes, including approvals, chargebacks, refunds, and reviews. It then weighs the risk of fraud against a "value" of each unique customer that in turn helps to identify legitimate transactions from fraudulent ones. The technology applies a "safety rating" to every transaction after comparing it to billions of data points in its global network and then, based on its risk, uses workflows to redirect a suspect transaction for further review or suspension.

"Fraud is an ever-evolving problem, so businesses must ensure their prevention technology evolves right alongside to maximize protection," Rich Stuppy, COO at Kount said in a recent press release. "Our solution is designed to be dynamic, providing both protection and opportunities to increase sales and revenue."

While Kount's management believes that its artificial intelligence-driven technology is superior to its competitors, the company is playing in an increasingly crowded field of tech companies that are also using machine learning to counter fraud, such as Accertify, CyberSource and Easy Solutions. All of these services provide various levels of fraud management and online authentication features that employ a combination of rules, metrics and geolocation services to help identify and investigate possible fraudulent transactions. Pricing for these services vary, but are mostly based on the volume of transactions.

This is all good news for online merchants - both big and small - who battle daily with chargebacks and potential liabilities caused by fraudulent transactions. As machine learning gets smarter and data sets become larger, the kinds of software and services offered by companies like Kount and others are quickly becoming a critical component for any ecommerce infrastructure. It's not just about saving money. Like PetSmart, it's about protecting your brand.

"No matter how the criminal acquired the credit card information to make a fraudulent charge, the customer will associate it with PetSmart and that PetSmart didn't protect their identity." Chad Evans, manager of e-commerce investigations at the company told ZDNet.