Back to Blog

The uptick in ecommerce has inspired many trends for businesses of all sizes and stripes —and not all of them are positive. While most omnichannel and ecommerce retailers have historically been concerned with chargeback fraud, some now recognize a type of fraud that has, quite possibly, even more impact on the bottom line.

Our work at Appriss® Retail has uncovered an alarming rise of ecommerce order claims and claims fraud and abuse, posing the threat of substantial losses to retailers if it’s not addressed. With the continued growth in ecommerce and increasing sophistication of fraudulent transactions (as well as honest consumers that have taken advantage of more lenient policies), it’s important for retailers to put measures in place to track and combat these practices.

When it comes to retail consumer fraud, most ecommerce and omnichannel retailers focus primarily on preventing chargeback-related fraud. However, with the growth in online shopping, ecommerce retailers are now recognizing that claims fraud may have even more impact on the bottom line than chargebacks.

Understanding Claims vs. Chargebacks

When it comes to retail consumer fraud, most ecommerce and omnichannel retailers focus primarily on preventing chargeback-related fraud. However, with the growth in online shopping resulting from the pandemic, ecommerce retailers are now recognizing that claims fraud may have even more impact on the bottom line than chargebacks. These losses are likely to continue or grow worse as the NRF predicts that non-store and online sales are expected to grow 11-13 percent in 2022 as consumers continue to utilize ecommerce, whereas total retail sales are only projected to increase between six and eight percent.

For purposes of this article, it’s important to first make the distinction between claims and chargebacks in terms of who is contacted to resolve the issue:

  • Claim: A consumer contacts the retailer to resolve an issue with an order’s merchandise or delivery. This includes claims that the item was not received, damaged, incomplete, or not what was expected.
  • Chargeback: A consumer contacts the credit card company or other payment provider and has the charge for an order reversed saying they did not make the purchase in the first place. There are two types of chargebacks, ones where the consumer legitimately did not make the purchase, and ones where the consumer made the purchases, but claims they did not.

Chargebacks were designed by payment providers, such as MasterCard, to protect consumers from charges made with stolen credit cards. Some consumers, however, use chargebacks in other situations, such as claiming not to have ordered of a package that was actually delivered. All chargebacks contain an element of fraud, whether it was the card number that was stolen, or a false chargeback was filed.

Claims present an added area of complexity since there are many legitimate claims that can be filed. When consumers’ orders are damaged during delivery, delivered late, damaged, or stolen, they turn to your customer service team for help. The shopper files a claim, and, if valid, the Customer Service Representative (CSR) makes an adjustment to the account such as offering an appeasement (basically a credit back to the customer) or reshipment according to company guidelines. These guidelines are designed to provide service with as little friction as possible for the shopper. Claims are an important component in consumer satisfaction.

Understanding the Impact of Claims Adjustments

Appriss Retail estimates total claims adjustments to be anywhere from two to four percent of all ecommerce sales depending on the retailer. In 2021, U.S. ecommerce sales totaled $1.050 trillion (according to an NRF report), which translated into $21-42 billion in appeasements and reshipments. In future years, retailers may face potentially greater losses as ecommerce grows and abusers and fraudsters become more sophisticated.

How Can You Calculate Your Claims Fraud?

As established above, valid claims and fraudulent claims combined are expected to be between two and four percent of ecommerce sales (though your own total claims may be different). It has been our experience that about 10 percent of those claims are fraudulent. For a quick estimate of your claims fraud situation, you can use this formula:

($ ecommerce sales x 2% claims) x 10% fraud = Estimated loss to claims fraud

The fraudulent claims are usually made by a small percentage of individuals who repeatedly make claims either under their own IDs or under multiple IDs. While it is perfectly understandable that something could go wrong during shipping that leads to a dissatisfied shopper, this is not likely to happen on a regular basis for most consumers.

Retailers typically do not have the analytics power to distinguish between valid claims and sophisticated scams, and therefore err on the side of caution and fulfill more claims than they should.

What Are Retailers Doing to Weed Out Claims Abuse and Fraud?

Claims abuse or fraud can take many forms. At a basic level, when a consumer makes a claim to a retailer that is untrue, with the intention of receiving a refund or additional item free of charge, this impacts the retailer negatively. These types of claims and associated adjustments (including financial appeasements and reshipments) may include:

  • The item was not received (send it again or compensate me)
  • The item was damaged or different/less than expected (send it again or compensate me)
  • The item came late (adjust/credit back the shipping charge)

For a claim to reach the level of fraud, it usually involves several purchases and potentially creating fake consumer accounts that make it difficult to trace the claims activity. CSRs are trained to help and want to satisfy the shopper, but this may not always result in smart decisions about whether a refund or reshipment is justified.

Retailers typically do not have the analytics power to distinguish between valid claims and sophisticated scams, and therefore err on the side of caution and fulfill more claims than they should.

How Can Retailers Find Sophisticated Scams Hiding Behind Fake IDs?

Detecting claims fraud requires specialized knowledge to create the appropriate analytical models that work in real-time to identify potentially fraudulent transactions based on purchasing patterns. However, this type of analysis is beyond the reach of most retail companies. Appriss® Retail has been in the business of helping leading retail chains worldwide reduce shrink, prevent return fraud, detect and remediate employee fraud, and improve operational processes for more than 20 years. We’ve honed our technology to discover the missing links among data points that can reveal commonalities in transactions that a retailer can use to identify claims fraud and abuse. Fraud is a very complicated problem to solve; however, the sheer complexity of our algorithm when applied to “connect the dots” among claims associated with particular IDs at scale can provide real-time authorization capabilities to service reps.

At Appriss Retail, we use Artificial Intelligence (AI) within a model-based approach to assess a number of risk factors (frequent returns of the same product, to the same address, or using the same credit card, etc.) that come together to create a risk measurement—a threshold above which a transaction might be identified as fraudulent. The retailer can choose whether to enforce the model at that level or to be more lenient or limiting. The primary objective of the system is to ensure that legitimate claims are recognized immediately, and the CSR can confidently make the appropriate adjustments.

Case Study: Millions in the Balance

Using our AI and analytics models, we found that for one omnichannel retailer, the 1.1 percent of consumers with the most post-order adjustments accounted for 10 percent of adjustment dollars. We also discovered that 0.3 percent of individuals were responsible for $2.8 million in adjustments. In other words, by denying these adjustments based on the relevant prior transaction history for specific linked IDs, this retailer could likely save close to $3 million annually. Upon further analysis, we discovered that at this retailer, $447,000 in total adjustments were requested by 25 individuals. By curbing this behavior, this retailer is also curtailing future fraud, thus limiting potential greater harm to the business as well.

Conclusion

Claims have always been regarded as a cost of doing business for retailers, but our research estimates that 10 percent of claims are fraudulent, and this number is likely to grow unless these behaviors are curbed. By applying AI and advanced analytics, retailers can potentially recoup millions of dollars while ensuring excellent customer service for consumers with legitimate claims.

In the end, it is up to each retailer to decide the threshold at which a consumer is denied a return transaction, and how far they will go to enforce their decisions should the consumer question their denial. But without some type of solution to alert them to suspicious activity, retailers are at the mercy of this unfortunate trend as they try to improve operations to meet the ever-expanding demand for ecommerce.

This article first appeared in The Robin Report.

Order Claims: A Growing Source of Ecommerce Fraud

READ NEXT

Author

David Speights, PhD, Chief Data Scientist, Appriss Inc.

David Speights, PhD, chief data scientist at Appriss, Inc., is the lead author of “Essentials of Modeling and Analytics: Retail Risk Management and Asset Protection” which was co-authored by Daniel M. Downs, PhD, and Adi Raz, DBA. He resides in California, USA, and speaks internationally on the intersection of artificial intelligence and retail.

Other Posts By This Author