3 Tech-Driven Solutions Turning the Tide Against eCommerce Fraud

People have been using the internet to buy things for almost as long as the internet has existed. The first known internet purchase happened way back in 1994 when entrepreneur Dan Kohn's NetMarket sold a CD to a Philadelphia man named Phil Brandenberger. That transaction was also the first to use encryption to protect a shopper's credit card data.

But it didn't take long before scammers and other ne'er-do-wells realized that stealing credit card numbers wasn't the only way they could take advantage of eCommerce websites. They instead turned to identity theft, chargeback fraud, and all manner of other methods of gaming the system.

And eCommerce sites have been engaged in a never-ending war against fraud ever since.

That war has reached a new level of intensity as the COVID-19 pandemic pushed more consumers to patronize online retailers. The spike in business created the necessary cover for a new wave of fraud attempts, which has driven up costs for retailers. At last count, every $1 of fraud now costs them roughly $.3.36.

But eCommerce sites are starting to fight back with a new range of technologies aimed at proving shoppers' identities, detecting potential fraud while it's in progress, and preventing unauthorized purchases before they happen. Here's a look at three tech-driven anti-fraud methods now going into service to fight eCommerce fraud today.

Tokenization to Limit Losses

The first line of defense that online retailers are turning to in large numbers is a new account security approach called tokenization. The most prominent purveyor of the technology is none other than credit card giant VISA. Their VISA Token Service provides an API that allows retailers to connect with the VISA transaction verification network without ever transmitting sensitive customer data.

Instead, the system relies on dynamically generated tokens that stand in for the actual account information and that are valid only for transactions with one specific retailer. This reduces the chance of the retailer suffering losses due to data breaches. And it also gives the retailer greater control over the transactions themselves. They can, for example, set purchase limits that trigger alerts, or even force tokens to expire after a preset number of transactions. The system is already so popular that VISA reports having generated over 2 billion tokens since its launch.

But VISA's not alone in trying to push tokenization. Third-party payment processors like Adyen have also made tokenization a key component of their security efforts. And their tokenization service is card-agnostic, meaning that retailers using the service gain the benefits of tokenization on every payment type they accept. It's an approach that is easier to implement for existing retailers and may become the de-facto transaction processing network standard soon.

New User Verification Methods

From the beginning, the essential challenge facing online retailers in the fight against fraud was finding ways to verify the identities of shoppers. In a card-not-present scenario, the only means of identifying a shopper used to be limited to the information tied to the credit card they were trying to use. But today, companies like Trust Swiftly are creating multi-layered identity verification systems that go much further.

In their case, retailers may choose from more than ten types of authentication methods and can trigger those methods using integrated risk-scoring. For example, low-risk transactions may proceed without verification, medium-risk transactions may trigger SMS ownership verification, and high-risk transactions could require submitting a selfie for biometric comparison against an on-file ID. And because retailers have control over the escalation, they can balance their anti-fraud needs against customer convenience in whichever way suits their needs.

Artificial Intelligence to Enable Real-Time Detection

Although still not common among smaller eCommerce operators, the use of sophisticated artificial intelligence (AI) solutions to detect fraud is growing. They owe their creation to the sheer volume of historical transaction data now available to serve as training for their algorithms. Using that data as a base of comparison, such systems can analyze transactions in real-time to look for the tell-tale signs of fraud.

That's just how the solution offered by Forter works, and it has attracted the largest participating merchant network in the industry. In their implementation, an anti-fraud AI will compare variables like past transaction histories, purchase locations (via GPS data), and purchase amounts against any new transactions on a given account. If the system deems a new transaction to be unusual or not in keeping with a consumer's tendencies, it can intervene as needed.

In most implementations, an anti-fraud AI will compare variables like past transaction histories, purchase locations (via GPS data), and purchase amounts against any new transactions on a given account. If the system deems a new transaction to be unusual or not in keeping with a consumer's tendencies, it can intervene as needed.

It's the same kind of approach that major card issuers like American Express have used to combat fraud for years now. In their specific case, an AI anti-fraud system increased their fraud resolution rate by 100% while reducing point of sale disruptions by 21%. And as the costs of AI development continue to drop, it's an approach that's certain to become a staple of eCommerce anti-fraud measures soon.

A Tech-Driven Problem with Tech-Driven Solutions

Right now, the incidence rate of fraud in the eCommerce space has never been higher. And yet, the prospects of reining it in have never been better. Through the increasing use of advanced identity verification and tokenization, online retailers are already making a big dent in the problem. And as more AI-powered solutions enter the picture, the situation will only improve.

And though it's unlikely that the problem of online transaction fraud will ever go away completely, it's safe to say the tide is now turning. It's a perfect example of a new generation of technology solving a problem created by its predecessors - continuing an inexorable march of progress that benefits all. Except for scammers and criminals, that is.