01 - The Acceleration of Return Abuse
02 - Why Return Abuse Outpaces Legacy Detection
03 - How Fragmentation Perpetuates Return Abuse
04 - Quantifying Your Return Abuse in 3 Steps
05 - Shaping Policy with an Identity-Based Approach
06 - How One Leading Fashion Retailer Decreased INR Abuse by 95.5%
How Leading Merchants Stay Ahead by Shifting from Incidents to Identity
07 - How to Evaluate an Effective Identity Network
08 - What Leading Merchants Do Differently
09 - How Nordstrom Saved $2.5 Million Last Year
10 - Assess the Impact on Your Revenue
The Ray-Ban x Meta Smart Glasses were one of last year’s hottest holiday gifts. Averaging $314 per pair, the smart glasses can take pictures, play music and calls, take voice commands, and much more. Countless pairs were returned over the past two months, a holiday gift inevitability. Now imagine 318 million of them were returned, totaling $100 billion.
That’s how much retailers really lost to return abuse last year. And that number is only going up, creating more and more revenue leakage. The Merchant Risk Council (MRC) found that over the past year, 57% of merchants reported increasing rates of refund and policy abuse, largely driven by false Item Not Received (INR) claims and attempted returns of used, damaged, or intentionally incorrect items. Gen AI has allowed abuse to scale so rapidly that it seems impossible to keep up. Merchants know it’s a problem. The National Retail Federation (NRF) found that 85% of merchants are using AI and machine learning in the returns process to identify and combat abuse. But only 38% believe these tools are truly effective.
Return abuse is an incredibly difficult problem to solve because there’s so much nuance. How can you distinguish a customer who regularly files false INR claims from someone whose packages keep getting stolen? Policy may treat them identically, even though one customer is a perpetrator and the other, a victim.
The Double-Edged Sword of Return Fees
Wardrobing, false INR claims, shipping back incomplete orders and empty boxes: all the classic forms of return abuse are still alive and well. Technology has simply made it easier for people to perpetuate these scams at a much higher rate.
With Gen AI, anyone can doctor a receipt or “damage an item,” creating a believable illusion of torn clothes, cracked cosmetics, or deformed laptop batteries. And they are. Forter data shows that AI-generated damage is the fastest growing form of return abuse. Sophisticated fraud rings have scaled their operations, in many cases, offering “returns as a service” in exchange for a percentage of the proceeds.
At the same time, return abuse has also become normalized among “regular” customers. NRF data shows that 62% of customers admit to abusing return policies. Additionally, Loop Returns found that 30% of customers who engage in wardrobing do so weekly.
The Acceleration of Return Abuse
increase in return abuse during Cyber Month 2025
Source: Forter
They need a full picture of who a customer is — their patterns and behaviors across many brands, not just the merchant’s own — in order to create the most effective policies. Policy and the models that enforce it should both be powered by identity intelligence: the real-time ability to stitch together thousands of first-party signals across devices, sessions, accounts, and behaviors into a single view. Without identity intelligence, fraudsters will continue to operate in the gray area, exploiting every loophole and creating more revenue leakage.
22%
merchants saw refund and policy abuse increase by at least 50% over the last year
Source: MRC
30%
year-over-year increase in false INR claims
Source: Forter, PwC
Why Return Abuse Outpaces Legacy Detection
Many merchants attempt to reduce return abuse with rules-based policy. Examples include introducing return fees or limiting the number of accepted returns. Nearly three-quarters of merchants have started charging for at least one return option in the past year.
The problem is, return abuse has so much gray area that black-and-white policy alone can’t solve.
Static rules don’t distinguish customers who commit INR fraud from those who are genuine victims of package theft, which 31% of Americans experienced last year, according to home security research firm SafeWise. Return policies also factor into many peoples’ shopping decisions. Some online shoppers have tried clothes on in-store beforehand, but most don’t. If something doesn’t fit, or even if the blue looked different on the website, customers want to know they can return it for free and without hassle. Rigid return policies inadvertently contribute to revenue leakage by alienating these customers before they’ve even had the chance to abuse them.
Implementing return fees has been positive for merchants….
Merchants must know the person behind the return.
…but there are also serious consequences
How Fragmentation Perpetuates
Return Abuse
Most abusive behaviors are invisible and not explicitly labeled “abuse.” Hidden costs are also spread around teams, none of which formally owns the return abuse problem.
have recouped revenue from returns
42%
53%
report a reduction in return rate
saw average order value decrease
34%
37%
have lost customers
Forter can help you uncover your undetected abuse and how much you can recover in lost profitability from returns and INR abuse.
Cost of goods
Logistics costs to process and ship items
Margin hits from refunding fraudulent transactions
Warehouse workers packing future return abuse orders
Customer service teams fielding false INR calls
Return abuse often straddles the line between fraud and customer experience. Customer service representatives process returns and issue refunds — but they’re not trained to identify fraud. Similarly, if a customer’s profile doesn’t include refund data, the fraud team may not be aware of serial abusers.
Everyone's Problem.
Nobody’s Responsibility.
TABLE OF CONTENTS
Payment fraud, account takeovers, organized fraud rings
Percentage of transactions or sales that are fraud-coded chargebacks, percentage of orders that are auto-approved or auto-declined
Speed, customer satisfaction
Responding to customer queries; processing orders and returns; resolving complaints, including returns
Fraud Team
Customer Service Team
Responsibilities
Primary KPIs
That fragmentation makes the true cost incredibly difficult to quantify, especially since no team exclusively owns “return abuse.”
“We calculated the singular metric for total cost of fraud, but we brought it down to an order level and that then allowed us to tie it back to all the different lines of businesses within Wayfair,” said Matt Lampert, Head of Fintech and Loyalty Data Science, at IMPACT 2025. “If I'm talking with a sales team and they're wondering why we're blocking these high value B2B orders, I can explain that not only is this the cost of fraud impacting your line of business, but here's how much time your agents are spending actually talking to fraudsters.”
Quantifying the impact begins with understanding the amount of abuse taking place. It can be challenging to assess because abusers often mask their identities. There are also many other numbers to consider beyond the underserved refund. When customers file false INR claims, you lose the item and the ability to earn a profit off it, in addition to various operational costs.
Wayfair has developed their own “total cost of fraud” (TCOF) metric centered around profitability, not just fraud. The number includes all costs, both direct (vendor fees, network fines, chargebacks) and indirect (overhead, false positives).
You can take a similar approach to calculate your own total cost of abuse.
Clearly define “abuse”
There’s no universal definition of “abuse.” The term varies from one business to the next, and even one team to the next. Merchants draw the line with wardrobing thresholds, INR claim frequency, and return-to-purchase ratio cutoffs, to name a few.
It’s important to have your own definition that spans the entire organization. That way, a customer service rep and a fraud analyst would classify the same behavior the same way.
1
3
Calculate your total cost of abuse per order, account, or program
Start with by calculating your total cost of abuse per order. Then roll up to account level to identify serial abusers. Lastly, segment by channel or campaign to find where the abuse is concentrated.
That will give you insight into where return abuse has the biggest impact and its downstream effect on the customer experience.
The Hidden Cost of Return Abuse
Request your return abuse assessment
10%
Cost of inventory + Revenue if you did sell the inventory to a non-abuser + Operating expenses associated with the return = Profitability impact
Policy alone can’t stop return abuse because it’s rooted in transactions. If a fraudster uses a revolving door of accounts, email addresses, and payment methods, they look like a new customer every time. Without a history of abuse, they look low-risk to the models that approve their return.
The vast majority of merchants are using AI and machine learning to identify and combat return abuse. However, only 38% believe those tools are truly effective. If the models are pattern-matching on transactions, decisions are based on what someone does, not who someone is.
Identity intelligence gives the models a complete picture of who a customer is: their previous interactions, prior refunds, INR claims, returns, loyalty promotion data, and behavioral data across the network. That visibility, in turn, allows merchants to create more effective, dynamic policies.
When a merchant knows who a customer is, they can confidently answer the question: Is this a trusted customer with an occasional issue? Or is it a serial abuser exploiting our policies?
When merchants understand their customers on the identity level, AI can evaluate their patterns for return abuse risk. From there, merchants can decide how they want to approach that risk and create policies accordingly.
Policy is ultimately only as strong as the identity intelligence powering it. Identity makes it possible for merchants to reduce abuse without punishing good customers, preserving the customer experience and their margins.
Trusted Customer: They have never displayed abusive behavior
Risk Level
Low
Return Decisioning
Approve instantly
Identity intelligence gives you confidence in your good customers. Because you know they’re low-risk, you can grant them faster approvals with no additional friction. The excellent customer service ensures they’ll keep coming back.
Abuser: They may have displayed abusive behavior on a small scale
Medium
Manually review, potentially allowing them to return with a caveat
When a customer service representative has visibility into identity, they can see whether the customer has a pattern of abusive return behavior, allowing them to make a more informed decision. They may issue a refund — with additional proof or in exchange for store credit only. But issuing a full refund doesn’t have to be the default.
Fraudster: They may be part of a coordinated refund or INR ring
High
Automatically deny
Identity intelligence doesn’t just flag one fraudulent refund in isolation. It can link that person to a coordinated ring operating across various accounts and channels. Their returns are outright denied and their claims never reach the customer service team’s already-long queue.
Quantifying Your Return Abuse in 3 Steps
Shaping Policy with an Identity-Based Approach
One high-growth fashion retailer built its customer experience on fast, trust-first issue resolution and frictionless post-purchase flows. That customer-centricity made the brand a prime target for return and INR abuse, creating millions of dollars in revenue leakage.
The brand initially tried to tackle the problem with strike-based policies, manual reviews, and thresholds on the number of claims, refund velocity, or spend. Because their tools treated each order and claim in isolation, abusers were able to adapt quickly, creating new accounts or tweaking their profile details. Using Forter’s rich identity intelligence, the brand connected behavior across accounts, devices, payment methods, and sessions. That allowed them to replatform INR control around identity, rather than transactions. With adaptable, identity-based policies, the brand automated “obvious” approval decisions, driving a 95.5% improvement in INR abuse incidents.
Success Story
How One Leading Fashion Retailer Decreased INR Abuse by 95.5%
A mediocre identity network is limited to a single region, vertical, or PSP. That vastly limits how many identities are included, not accounting for someone who commits abuse with retail brands and travel merchants. The average American also has three or four credit cards. The average fraudster is likely to have more.
How to Evaluate an Effective Identity Network
Business Diversity
The best identity networks cover billions of unique identities across businesses and industries at enterprise scale. That way, “new” shoppers to one merchant are already known. Someone may shop with a specific brand once a year, but their activity across other brands creates a clearer picture for the entire network.
First-Party Data
Many identity networks are essentially wrappers around a third-party feed. That’s risky because third-party data is inherently unreliable. If an external source changes or third-party data feed suddenly stops, decisions break. Merchants need first-party data from real commerce interactions: sign-up, login, and checkout.
Cross-Journey
Identity intelligence should go beyond checkout, and cover sign-up, login, payment routing, returns, and disputes. That way, the identity graph supports both fraud and abuse decisions from the same spine. Ideally, the network can also power-real time decisions for abuse across the journey. Enrichment-only networks still force merchants to maintain complex rules or additional tools.
Rich Profiles
The more attributes an identity has, the easier it is to stop a serial abuser from perpetuating scams at multiple brands within the network.
High-Quality Linking
Beware networks that stitch together static attributes, such as IP, email, and device ID. The most effective networks consist of person-level identity graphs. Probabilistic linking connects records from different datasets and calculates whether they’re likely to be the same person. Connecting emails, devices, cards, addresses, and sessions, merchants can better identify a fraudster, even when they switch up details. If every email and device combination is treated like a unique individual, it’s impossible to spot patterns.
Network Freshness
The most effective networks don’t just provide clarity into who customers are. They also remember them. Short-window networks, which only account for one or two weeks of activity, miss slow-burn abuse. Persistent histories will flag long-cycle resellers or seasonal return abusers, for example.
Rich Profiles Have Thousands of Attributes
2
Build your total cost of abuse
The goal is a single per-order cost of abuse figure: One number that lets you tie return abuse back to specific business lines, justify policy decisions internally, and quantify the recovery opportunity. Most merchants undercount significantly because they only capture the direct refund. The real cost has two components: what you pay out, and what you fail to earn.
Cost
What's Often Missed
Refunds, credits, and discounts
Courtesy discounts issued to retain a complaining customer, which are often processed outside the standard returns flow and never attributed to abuse
Cost of goods
The full landed cost — not just the wholesale price
Reverse logistics
Return label, outbound shipping on the original order (especially if free or refunded), packaging, and per-return fees from your reverse logistics provider or returns management platform
Restocking and warehouse labor
Labor cost per return processed (inspection, sorting, disposition), not just items that result in a write-off
Abused promotions, loyalty points, and gift cards
Points earned on purchases that are subsequently returned and not clawed back; promotional discounts on orders returned after the promotion window closes
Third-party vendor and chargeback fees
Per-transaction fees from fraud or returns vendors, which compound quickly at volume and are frequently buried in SaaS budgets
Customer service volume
Agents handling false INR claims, who spend significantly more time per interaction due to evidence review and escalation
False positives
Good customers incorrectly flagged or denied who churn — your most underestimated indirect cost because it shows up as lost revenue, not an expense
Free replacements
Full landed cost of the replacement unit plus outbound shipping, on claims that were never investigated
Lower CSAT and increased churn
Return experience as a driver of dissatisfaction in your VoC data; churn rate among customers who were friction-gated or denied versus your baseline
Lower CLV from over-restriction
Legitimate high-return customers — parents buying children's clothing in multiple sizes, for example — whose long-term value is being written off alongside genuine abusers
What Leading Merchants Do Differently
Dynamic, identity-based policy, with frictionless experiences for good customers
Leading Merchants
Legacy Merchants
Static, rules-based policy that treats every customer the same
Automated decisions with human intervention reserved for gray area
Case-by-case manual reviews
Identity intelligence that spans geographies and businesses
Data specific to a single merchant, industry, or geography
Cross-journey, omnichannel visibility that covers in-store and online
Siloed, channel-specific decisions
Nordstrom wrote the book on customer service. Literally. The Nordstrom Way to Customer Service Excellence demonstrates how seriously Nordstrom takes the customer experience. Policies like the gift with purchase program were very attractive to both loyal customers and fraudulent resellers — two groups Nordstrom had a difficult time telling apart.
Working with Forter, Salesforce, and AWS, Nordstrom is now better able to understand who’s who. Blocking abuse from likely resellers saved Nordstrom $2.5 million in revenue last year. Rather than conducting manual reviews and addressing fraud malfunctions, Nordstrom’s fraud team can now focus their time and energy on deconstructing siloes and redesigning strategies.
SUCESS STORY
How Nordstrom Saved $2.5 Million Last Year
“Forter has freed up our time and capacity to focus on strategy — not firefighting. With Forter, we can harness all our data and act with confidence to protect the customer experience on all fronts.”
Christine Deibel, Senior Program Manager, Fraud
READ THE REPORT
Assess the Impact on Your Revenue
Totaling $100 billion last year, return abuse plagues nearly every merchant. It’s not a matter of “if” it’s a problem. The real question is, how much of a problem is it? As technology makes abuse easier to perpetuate, that number is primed to grow — but it doesn’t have to.
The right tools can see the full picture. Once merchants dig in to discover the total cost of abuse, they’re surprised by what they uncover.
Email addresses
Shipping and billing addresses
Payment methods
IP addresses
Device fingerprints
Merchant account IDs
Core identity and device
Browsing time
Purchase frequency Categories ordered Device, email, and card switch rates INR counts Returns and cancellations
Behavioral and history
Stolen card probability
Account takeover probability
Reputation level Number and type of chargebacks Fraud alerts Manual decisions
Reputation and risk
Request a return abuse assessment
RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE
RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE
RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE
RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE RETURN ABUSE
Do you know how much return abuse is really costing you?
Return Abuse