Topic starter
11/07/2026 2:01 pm
TL;DR: Ecommerce returns are shifting from a customer-experience issue into a fraud and operations control problem, with Signifyd citing research that returns cost merchants about $850 billion in 2025 and 9% of that activity is tied to fraud. Agentic returns abuse raises the stakes because automation can scale repeat claims faster than legacy review workflows can separate good customers from coordinated abuse.
Editorial analysis by NHI Mgmt Group, based on content published by Signifyd: “The complete guide to ecommerce return optimization in 2026”.
By the numbers:
- Returns cost merchants about $850 billion in 2025, according to the National Retail Federation and Happy Returns research cited by Signifyd.
- 9% of returns activity is tied to fraud, according to the National Retail Federation and Happy Returns research cited by Signifyd.
- The cost of processing an online return averages 21% of an order’s value, according to a Pitney Bowes BOXpoll survey cited by Signifyd.
Key questions
Q: What breaks when ecommerce returns are treated as a simple service process?
A: The main failure is that fraud, abuse, and customer friction are handled with the same rules.
Q: Why does agentic returns abuse increase fraud risk for merchants?
A: Agentic returns abuse increases risk because automation can generate many return requests across multiple accounts faster than manual review can distinguish normal activity from coordinated abuse.
Q: How can teams tell whether return friction is working?
A: Return friction is working when it reduces abusive behaviour without materially increasing complaints, abandonment, or false declines among legitimate customers.
Practitioner guidance
- Build a risk-based returns policy Create separate handling rules for low-risk and high-risk returns so legitimate customers get fast resolution while suspicious patterns face inspection, verification, or delayed refunding.
- Use returns data to find root causes Track return reasons, timing, product categories, and customer behaviour to tell the difference between policy abuse, sizing issues, product quality problems, and repeated fraud patterns.
- Add targeted friction at the refund decision Require step-up checks for item swapping, empty-box claims, repeated wardrobing, or high-velocity serial returns instead of applying the same controls to every request.
Bottom line: Ecommerce returns now sit at the intersection of customer experience, fraud control, and margin protection, so policy design has become a governance problem.
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
11/07/2026 2:11 pm
Returns governance is becoming a fraud-control discipline, not a customer-service side function. The article shows that return workflows now influence margin, inventory integrity, and abuse exposure at the same time. That makes returns data, refund decisioning, and policy design part of a broader control framework rather than an isolated ecommerce operation. Practitioners should treat returns as a governed business process with fraud consequences, not a back-office convenience layer.
A question worth separating out:
Q: When should merchants require inspection before issuing a refund?
A: Require inspection when the item is high value, prone to substitution or tampering, or associated with repeated abuse from the same customer or account cluster. Inspection is most useful when physical condition must be confirmed before money leaves the merchant.
👉 Read our full editorial: Ecommerce return optimization is becoming a fraud control problem