MAG Insights

Announcements from the MAG & Featured Articles

MAG Sponsor Spotlight: 7 Trends Hitting Retail Fraud Now (MAG Quarterly- Volume Five, Issue Three)

By Coby Montoya, Product Manager, Accertify, Inc.

September 7, 2017

As a retail fraud manager, you face a tough balancing act: detecting fraud while supporting a superior customer experience. Every retailer wants to execute at lightning speed—but that gives your team less time to make risk decisions. And fraud is rising, along with the cost: LexisNexis reports the cost per dollar of fraud rose from $2.23 in 2015 to $2.40 in 2016.[i] Still the biggest blow may come from turning away good customers with high lifetime values.

To respond quickly to complex and growing threats, you need to know what’s coming. Here’s a look at seven fraud trends and a tip on how to combat each one.

#1 EMV-driven online fraud: With chip cards at POS, the path of least resistance for fraudsters has shifted to CNP transactions online. In the six months after EMV implementation, online fraud grew 11 percent.[ii]

Tip: Monitor trends as fraudulent activities happen, not when chargebacks come in. A holistic digital-fraud strategy incorporates authorization data, a robust rules engine, machine learning, link analysis, and third-party providers who validate attributes of risky orders.  

#2 Account takeover/loyalty fraud: Although payment fraud has been the focus for years, hackers and thieves may find loyalty accounts more tempting due to lower security. Fraudsters can take over accounts via sophisticated phishing techniques or with credentials acquired on the dark web.

Tip: Extend your fraud strategy to all touch points. You’ll be better able to understand your customers’ true behaviors and to detect account compromise and malicious activities.  

#3 Returns fraud: Retailers face significant dollars lost when customers return high-end merchandise they used or stole. Even fraudsters who do this habitually may be difficult to detect without the right tracking mechanisms.

Tip: Ensure that your returns processes are not managed in silos. This data ideally should flow into your fraud platform. If this is not possible, have a process in place to identify repeat offenders and abusive behaviors.

#4 Omni-channel fraud: Multiple sales channels—in-store, phone, online—can move a card-present transaction to a CNP one. Without communication and tracking across channels, fraudsters can bypass protections. For example, in-store pickup makes the delivery address moot. Fraudsters will try a transaction with a card, then place dozens of orders if it works.

Tip: Tailor your fraud strategy to specific channels. In-store pickup, for example, makes delivery address velocity rules moot. Customized strategies will minimize disruption to good customers.

#5 Mobile/native app fraud: Mobile commerce is growing rapidly, but it’s critical to distinguish between web and native app mobile commerce. Each provides different data elements, and a customized fraud strategy will reduce friction and minimize risk.

Tip: Understand the nuances that characterize mobile web and native app commerce. Deploy device SDK capabilities for native apps, and customize mobile web fraud detection strategies to differentiate them from desktop strategies.

#6 One-click checkout: One-click checkout may be nirvana for marketers and customers, but a challenge for you. There are two main data types in our world: active and passive. Your customers insert active data, such as name, card number, and email, in a screen. But unknown to customers, you can also collect passive data, such as IP address or device ID, for risk decisions. For a secure, one-click experience, don’t short yourself on data for fraud screening.

Tip: Rely more heavily on passive data for risk decisioning to support fast, frictionless payment.

#7 Automated fraud: Fraudsters today can download sophisticated tools from the black market to perform technically complex tasks efficiently. Scripted attacks that can be carried out quickly may result in large losses.

Tip: Analyze user behavior and model the difference between transactions entered by a human and by a bot. A bot may have atypical cursor and mouse movement, for example.

[i] 2016 LexisNexis True Cost of Fraud Study,” LexisNexis, May 2016.

[ii] Andrew Meola, “Online fraud attacks in the U.S. are growing at an alarming rate,” Business Insider, April 20, 2016.