Stopping e-commerce returns fraud in its tracks
Stopping E-Commerce Returns Fraud in its Tracks
Returns fraud is incredibly hard to detect with the limited data marketplaces have. There was no consortium database to spot a “returns abuser.”
Our Sardine X consortium is adding returns abuse signals, and we’re talking to marketplaces representing double-digit percentages of online commerce.
Return Fraud Categories
There are countless typologies for fraud returns. Here are some examples:
- Wardrobing: Buying and wearing once only to return. Often considered the “victimless crime,” this is especially prevalent with fashion and clothing.
- Concession/refund abuse: Contact customer support to return a low-value item, knowing they’ll get a refund or concession because it’s too expensive for the merchant to manage the return.
- Using stolen cards: Buying a product with a stolen card to claim the refund into an account the fraudster controls
- Bricking: Stripping an electronic item for parts and then returning it
- Empty box fraud: Customers claim they received an “empty box”
- Receipt fraud: Re-using or creating a fake receipt
Fraudsters are creative. They will use online chat to speak to customer support. One scammer even filled boxes with dirt to the “correct item weight” to fool automatic returns processes.
Marketplaces are especially vulnerable to Returns Fraud
Because e-commerce merchants lack a physical store presence, ensuring customers can return goods (returns) is essential. To gain the confidence of shoppers, e-commerce platforms offer generous returns policies.
Sadly, this is often abused, especially since the pandemic.
With over $550 billion in e-commerce sales in 2022, returns fraud has become a massive problem. The National Retail Federation estimates that 7-10% of returns are fraudulent, costing U.S. retailers $18 billion annually.
Sophisticated fraud rings are exploiting return policies to launder money and steal inventory.
Even individual opportunistic fraud is rampant, with practices like “wardrobing” costing retailers heavily.
At Sardine, we take a data-driven approach to detect and prevent returns fraud across the user journey. Read on to learn how our fraud platform gives retailers the visibility and control needed to stop abuse.
The Rising Impact of Returns Fraud
While most returns are legitimate, fraud is undeniably on the rise:
- Return fraud costs U.S. retailers $18 billion per year, according to NRF estimates
- Fraudulent returns make up 7-10% of all returns, meaning billions in preventable losses
- 25% of annual returns happen between Thanksgiving and New Year's Day
- E-commerce fraud attempt rates during 2022 holiday season rose 82% globally
Empty box scams, stolen merchandise returns, wardrobing, and organized retail crime rings all plague retailers. And consumer perception that return fraud is a "victimless crime" enables the problem.
Tightening return policies risks reputational damage and lost sales, too. Finding the right balance requires a nuanced data-driven approach.
Generative AI makes marketplaces more vulnerable
Fraudsters already use generative AI to scale up scams, fraud, and money laundering. This exacerbates an already exponential growth in returns fraud.
With Generative AI, fraudsters can
Quickly create believable receipts and text at scale
- Automatically manage chatbot conversations
- Create deep fakes for customer verification for more high-value scams
To read more about how fraudsters use Generative AI and how we detect it click here.
The Challenges of Spotting Returns Fraud
- At account opening, data is limited: Marketplaces have never seen a user until they register. They’d rather err on caution and let that user register than block them for fear of high false positive rates.
- Focussing on just account opening or transactions: Legacy fraud tools focus narrowly on single transactions or account opening. This leaves blind spots across the user journey, enabling returns fraud to slip through.
- Friendly fraudsters look like good customers: Some schemes also fall into a gray area that static rules miss. An honest customer who wardrobes occasionally is still abusing the system. Transaction monitoring solutions might miss this.
During peak seasons like the holidays, waves of returns easily overwhelm manual review.
The patterns are easy to spot if you have enough data and the right data, but often, it's too late once abuse has happened.
The Sardine Approach
At Sardine, our intrinsic fraud prevention platform is purpose-built to overcome these challenges. We stop returns fraud across the entire account lifecycle with:
- Visibility into cross-account abuse through the SardineX consortium
- Sophisticated ML detecting organized rings and dynamic risk patterns
- Robust device fingerprinting to link accounts and identify returning devices
- Behavioral biometrics baselining normal activity and flagging anomalies
- Precise configurable rules tuned to your unique fraud vectors
- Identify and separate friendly fraud from compromised accounts
Using behavioral signals like changes in typing cadence or swipe patterns, our AI spots the difference between honest mistakes and deceptive fraud.
We can identify abnormal return volumes, link accounts to fraudulent devices, and recognize organized patterns across your customer base. All in real-time, before further abuse occurs.
Bottom line - we make the connections legacy tools miss while minimizing false positives that damage conversion and reputation.
E-commerce returns fraud is a $18 billion problem requiring specialized expertise and technology.
Get in touch if you want the ultimate database on returns fraud abuse. Join Visa, Experian and the worlds largest banks and Fintech companies to stop fraud before it happens.
Our team has decades of experience stopping fraud faster than it can scale.