Resources
Learn how abuse works, and what it costs
Writing, conversations, research and free tools from the Pinch team on return, claims and checkout abuse.
01
Field notes on fraud, abuse and the economics of returns.
02
In A Pinch: conversations with operators and investigators.
03
Guides, infographics and videos. Free to download.
04
Estimate your losses, check your policy, scan for abuse.

Engineering
·
Pandas to DuckDB: A Lightweight Switch That Took Our Feature Runs From Hours to Minutes
From 5–6 hour Pandas runs to under 10 minutes in DuckDB - without heavy infra. Notes from the trenches.
Pinch AI team

Product
·
The Missing Link in Returns Fraud Prevention: Disposition Data
Disposition data reveals the ground truth about what customers actually return, but it arrives too late to prevent fraud in real-time - so effective fraud prevention relies on pre-return behavioral signals that get smarter when validated by disposition data after the fact.
Chirag Vaya

Company news
·
Announcing Our Seed Round to modernize E-commerce risk and fraud infrastructure
Pinch announces it's seed fundraise of $5M to build the intelligence layer for commerce risk, modernize post-purchase risk and fraud infrastructure, and build the next layer of trust in digital commerce.
Arthi Rajan

EP
6
·
Michael Rasile
Michael Rasile, touring drummer turned fan-experience leader at Front Row and partner at Ceremony of Roses, joins Arthi and Jayan to talk about what really happens inside the war room when a major artist drop goes live. He walks through the Harry Styles drop that "looked like a roulette machine," the trade-off between blocking freight forwarders and protecting Billboard chart eligibility, why the line between a super-fan and a fraudster is razor-thin, and the under-discussed rise in chargeback abuse he believes is being fueled by AI cynicism toward the internet.

EP
5
·
Michael Vergara
When is fraud really abuse, and when is abuse just a customer taking advantage of a policy you left open? In our latest In a Pinch episode, Mike Vergara (veteran risk leader across RSA, PayPal, Blackhawk, and now Trustly) makes the case for a product-first approach to risk, unpacks the rise of first-party abuse, and shares the origin story of "human-in-the-loop" at PayPal a decade before the term existed. A must-listen for anyone building in fraud, risk, product, or fintech.

EP
4
·
Philipp Barthold
We sit down with Philipp, CTO of Lux Experience, the parent company behind iconic luxury fashion retailers Mytheresa, Net-a-Porter, and Mr. Porter. Philipp shares his journey from building neural networks for payment fraud detection in 2004 to running technology for one of the world's largest luxury e-commerce operations. The conversation dives deep into the unique fraud challenges of high-end retail (including a wild fake FedEx truck heist).

Infographic
·
Return Abuse Trends - 2025
Download the infographic highlighting key 2025 return-abuse trends from Pinch AI’s retail data, showing the widening gap between fraud and loyalty in e-commerce returns.

Video
·
Inside Return Abuse: How Shoppers are Gaming the System
Return abuse is costing retailers billions, yet most shoppers have no idea it’s happening. In this video, we dive into the growing crisis of return fraud and abuse, breaking down just how massive the problem really is, and the various methods bad actors use to exploit return policies.

Guide
·
Guide: Understanding E-commerce Return Abuse
This comprehensive guide provides an in-depth look at the various types of return abuse in e-commerce, their impact on businesses, and effective strategies for prevention. Learn how to identify and address fraudulent return activities to protect your bottom line and maintain customer satisfaction.
Free lookback · for qualified brands
See what you lost last year. Free.
01
Share 12 months of data.
Install the Shopify app or SFCC cartridge, or send order, return and claim exports.
02
Get your lookback in under two weeks.
What you lost to return and claims abuse, and who took it.
03
Qualified brands try Pinch free for three months.
Run it on live orders, then decide.