Ecommerce Return & Refund Fraud Statistics (2026): 30+ Data Points on Fraud Rates, Costs, and the Tactics Behind Them
Discover the latest ecommerce return and refund fraud statistics for 2026. Explore over 30 data points on fraud rates, costs, and tactics used by fraudsters.
Try it on your next invoice
Draft from text or voice, edit every field, and export a PDFβfree on the homepage.
Try AI invoice9% of all retail returns are fraudulent, which works out to roughly $76 billion lost to return and refund fraud in 2025 (National Retail Federation, 2025 Retail Returns Landscape). Returns themselves are a massive cost of doing business online β Americans sent back an estimated $849.9 billion in merchandise in 2025 β but the slice that is deliberately dishonest is what turns a margin problem into a security problem. Online channels make it worse: e-commerce purchases are returned at 19.3%, well above the 15.8% all-channel rate, because shoppers can't inspect goods before buying and increasingly treat their living rooms as fitting rooms. (This topic is often searched as "2023 ecommerce return and refund fraud statistics," but the figures here reflect the most recent NRF and Appriss Retail data available.)
The retailer response has shifted from writing off shrink to actively policing it, with the large majority now deploying AI to flag suspicious returns before refunds go out. We aggregated data from the National Retail Federation, Appriss Retail, and Adobe Analytics to quantify how big return and refund fraud has become, how it's done, and what's being done about it.
Key Takeaways
- 9% of all returns are fraudulent (NRF, 2025).
- That implies roughly $76 billion in fraudulent returns in 2025 (9% Γ $849.9B) (NRF, 2025).
- Total US returns reached $849.9 billion in 2025 (NRF, 2025).
- The online return rate is 19.3%, versus 15.8% all-channel (NRF, 2025).
- Appriss Retail estimates omnichannel return fraud is a ~$19 billion profitability leak (Appriss Retail, 2025).
- 71% of retailers that track it saw a rise in overstated-quantity returns (NRF, 2025).
- 65% reported a rise in empty-box or "box of rocks" returns (NRF, 2025).
- 85% of retailers are using AI to detect or prevent return fraud (NRF, 2025).
The Scale of Returns
Return fraud only makes sense in the context of how enormous legitimate returns have become, because fraud hides inside that flow. US shoppers returned an estimated $849.9 billion in merchandise in 2025, equal to 15.8% of all retail sales (NRF, 2025 Retail Returns Landscape). Online sales drive a disproportionate share of that volume: e-commerce returns run at 19.3%, materially higher than brick-and-mortar because the buy-first-inspect-later nature of online shopping all but guarantees a higher mismatch rate. The sheer size of the return river is what gives fraud its cover β when hundreds of billions of dollars in goods are moving backward through the supply chain, a dishonest 9% is easy to lose in the current. For merchants running high return volumes, the back-office reality is a constant stream of credit notes and re-bills, which is part of why scaling sellers lean on tools covered in our ecommerce statistics roundup.
| Metric | Value | Source |
|---|---|---|
| Total US returns (2025) | $849.9 billion | NRF, 2025 |
| All-channel return rate | 15.8% | NRF, 2025 |
| Online return rate | 19.3% | NRF, 2025 |
| Total US returns (2024, prior year) | $685 billion | NRF, 2025 |
2. Return & Refund Fraud Rate & Cost
Stripping out honest returns leaves a stubborn core of fraud that scales with the total. At 9% of returns, fraud accounted for roughly $76 billion in 2025 β calculated directly as 9% of the $849.9 billion in total returns (NRF, 2025). It is worth being precise about year-over-year comparisons here, because the methodology shifts: in 2024, NRF data pointed to a higher fraud share (around 15% of a smaller $685 billion return base, or roughly $103 billion), so the apparent "drop" reflects changes in how fraud is estimated as much as any real decline. A separate lens from Appriss Retail frames the same problem as a ~$19 billion omnichannel "profitability leak," a narrower figure focused on the directly recoverable losses retailers can attribute to fraudulent and abusive returns. However it's sliced, the takeaway is consistent: for every $100 of merchandise returned, close to $9 is fraudulent, and that comes straight off the bottom line.
| Metric | Value | Source |
|---|---|---|
| Share of returns that are fraudulent | 9% | NRF, 2025 |
| Estimated fraudulent returns (2025) | ~$76 billion | NRF, 2025 |
| Prior-year fraud estimate (2024) | ~$103 billion (~15% of returns) | NRF, 2025 |
| Omnichannel return-fraud leak | ~$19 billion | Appriss Retail, 2025 |
| Fraud per $100 returned | ~$9 | NRF, 2025 |
Caveat: 2024 and 2025 fraud figures are not directly comparable β NRF's estimation methodology and return-rate base changed between the two reports.
The Tactics: How Return Fraud Happens
Return fraud is not one scheme but a family of them, and retailers report the cruder physical tactics rising fastest. 71% of retailers that track these incidents reported an increase in overstated-quantity returns β claiming to send back more than was actually shipped (NRF, 2025), followed by empty-box or "box of rocks" returns (65%) and decoy returns using counterfeit or substituted items (64%). Beyond these, three behavioral schemes dominate the online world: wardrobing, where a customer buys an item, uses it once, and returns it; bracketing, where a shopper orders several sizes or colors intending to return most; and receipt fraud, where falsified or reused receipts are used to extract refunds or store credit. The pattern across all of them is that generous, frictionless return policies β the same policies that win sales β also widen the attack surface, forcing retailers to balance conversion against loss.
| Metric | Value | Source |
|---|---|---|
| Rise in overstated-quantity returns | 71% | NRF, 2025 |
| Rise in empty-box / "box of rocks" returns | 65% | NRF, 2025 |
| Rise in decoy / counterfeit returns | 64% | NRF, 2025 |
| Common online schemes | Wardrobing, bracketing, receipt fraud | NRF, 2025 |
4. Ecommerce-Specific Return Behavior
Online retail doesn't just have more returns β it has a structurally different return culture that feeds abuse. The 19.3% online return rate sits well above the all-channel average, and 82% of shoppers say free returns are a major factor in where they buy (NRF, 2025), which is exactly the incentive that fuels bracketing and casual over-ordering. Younger shoppers drive much of the volume, with Gen Z averaging 7.7 online returns over a twelve-month period β more than any other generation. The economics are circular: free, easy returns lift conversion and basket size, but they also normalize buying with the intent to send most of it back, and a fraction of that behavior crosses into outright fraud. For multi-channel sellers reconciling refunds and credit notes across storefronts, the operational load is real, and high-volume operations often manage it with a bulk invoice generator.
| Metric | Value | Source |
|---|---|---|
| Online return rate | 19.3% | NRF, 2025 |
| All-channel return rate | 15.8% | NRF, 2025 |
| Shoppers citing free returns as a major factor | 82% | NRF, 2025 |
| Gen Z average online returns (12 months) | 7.7 | NRF, 2025 |
5. Detection & Prevention
Retailers have moved from absorbing return fraud as a cost of doing business to fighting it with technology and tighter rules. 85% of retailers now use AI to detect or prevent return fraud (NRF, 2025), using it to score returns in real time, flag serial abusers, and catch patterns like repeated empty-box claims that human staff would miss. Many are also reworking policies β shortening return windows, requiring receipts or accounts, charging restocking or return-shipping fees, and steering refunds to store credit β though each tightening risks dampening the conversion that liberal returns were designed to boost. The strategic tension defines the category: every dollar saved from fraud prevention has to be weighed against the sales and loyalty that frictionless returns generate. Sellers operating across several channels and client accounts increasingly centralize this reconciliation, which is where multi-client invoicing fits.
| Metric | Value | Source |
|---|---|---|
| Retailers using AI to fight return fraud | 85% | NRF, 2025 |
| Rise in overstated-quantity fraud (tracking retailers) | 71% | NRF, 2025 |
| Common policy responses | Shorter windows, fees, store credit | NRF, 2025 |
Return & Refund Fraud by the Numbers
| Metric | Value | Source |
|---|---|---|
| Share of returns that are fraudulent | 9% | NRF, 2025 |
| Estimated fraudulent returns (2025) | ~$76 billion | NRF, 2025 |
| Total US returns (2025) | $849.9 billion | NRF, 2025 |
| All-channel return rate | 15.8% | NRF, 2025 |
| Online return rate | 19.3% | NRF, 2025 |
| Omnichannel return-fraud leak | ~$19 billion | Appriss Retail, 2025 |
| Rise in overstated-quantity returns | 71% | NRF, 2025 |
| Rise in empty-box returns | 65% | NRF, 2025 |
| Rise in decoy/counterfeit returns | 64% | NRF, 2025 |
| Shoppers citing free returns as major factor | 82% | NRF, 2025 |
| Gen Z average online returns/year | 7.7 | NRF, 2025 |
| Retailers using AI vs return fraud | 85% | NRF, 2025 |
Methodology and Sources
Every statistic on this page is traced to a primary source and linked inline. The fraud dollar value is derived transparently as the fraud share (9%) multiplied by total returns ($849.9B). We explicitly flag that 2024 and 2025 fraud estimates are not directly comparable due to changes in NRF's methodology and return base. This topic is frequently searched with a "2023" qualifier; we use the most recent NRF and Appriss data and label the baseline year accordingly.
Primary sources used:
- National Retail Federation β 2025 Retail Returns Landscape
- NRF β Returns Press Release
- Appriss Retail β Omnichannel Return Fraud
- Adobe Analytics β Holiday & Returns Insights
Recency: figures reflect 2024β2025 NRF and Appriss reporting. Any figure older than two years is flagged as "most recent available."
Last updated: June 2026. We update this page quarterly.
Ready to create your next invoice?
Use AI drafting on the homepage or sign up for a free account with cloud save and monthly plan limits.
Related articles
Atlanta Retail Small Business Statistics (2026): 40+ Data Points on the Local Retail Economy, Ownership, and Growth
Discover key insights into Atlanta's retail landscape with 40+ statistics on small business ownership, growth trends, and the local economy for 2026.
AI in Ecommerce Statistics (2026): 45+ Data Points on Adoption, Personalization, and Conversational Commerce
Discover key statistics on AI in eCommerce for 2026, covering adoption rates, personalization trends, and the rise of conversational commerce.
Medical Billing Industry Statistics (2026): 40+ Data Points on Market Size, Claim Denials, and the Cost of Getting Paid
Discover essential statistics on the medical billing industry for 2026, including market size, claim denials, and payment costs. Stay informed and optimize your strategy.
Remote Work Statistics (2026): 40+ Data Points on Adoption, Hybrid vs Remote, and the Return-to-Office Debate
Discover the latest remote work statistics for 2026, featuring over 40 data points on adoption trends, hybrid vs remote models, and the return-to-office debate.