Recent activityJust drafted a letter for a Visa 10.4 dispute against a returning subscription customer.

Chargeback Index.

Succinct, practical articles from experts designed to provide you with actionable insights to win.

3-D Secure Does Not Always Shift Liability, Even When It Was Invoked

3DS is a partial shield with specific shapes. The carve-outs cluster around what counts as a real authentication and around reason codes that 3DS never covered at all.

AI Has Entered the Chargeback Space on Both Sides and Has Not Changed Who Wins

Volume of AI-generated content in disputes has roughly doubled in a year. Issuers still review the cases by hand, and they have learned to dismiss AI letters faster than the templates they replaced.

13.2 Cancelled Recurring Is a Cancellation-Flow Failure

The dispute is a he-said/she-said. Whichever side wins, the underlying problem is the same, and it sits with the merchant.

The Breakeven on a Dispute Analyst Is Lower Than Most Merchants Assume

At a fully-loaded $50 an hour and 30 minutes per careful representment, the in-house cross-over sits around 60 disputes a month. Most growing merchants pass it without realising.

10.4 Is the Most Expensive Reason Code, and CE 3.0's Only Target

Card-absent fraud carries the highest representment cost and the lowest naive win rate. CE 3.0 is the only framework that helps, and it helps a narrow slice of merchants.

The Letter Is the Most Important Thing in Chargebacks, and Most Services Hide It

Auto-submit platforms, AI-rebuttal vendors, and processor dashboards all send a document the merchant never reads. A merchant who cannot see the letter cannot improve the letter.

Mastercard's Dispute Regime Differs From Visa's in Three Ways That Cost Merchants Cases

A Mastercard 4853 is not a Visa 13.1 with a different number. The rules differ on timing, on reason-code structure, and on workflow, and merchants who do not notice lose cases they should have won.

The Issuer's Analyst Reads Your Letter in 90 Seconds

One tired analyst, a structured scoring sheet, a queue of similar cases behind. The letter that wins is written for that workflow.

13.3 Is a Customer-Expectation Dispute, Not a Quality Dispute

Visa 13.3 turns on whether what the merchant promised matched what the customer ordered. Photo evidence of the delivered item is rarely as useful as the product description on the page where the customer clicked buy.

Digital Goods Disputes Turn on Access Logs, Not on Delivery

The issuer's question is not whether the digital goods were delivered. It is whether the cardholder used them after paying.

Marketplace Chargebacks Fall on the Wrong Party by Design

Marketplaces take the customer relationship and the dispute interface. They leave the chargeback liability with the seller. The structure is deliberate.

Services Businesses Need a Different Kind of Fulfilment Record

There is no tracking number on a haircut. The records that win services chargebacks are operational, not logistical, and most merchants do not capture them.

Why Subscription Chargebacks Defeat Pure-Play Dispute Services

Subscription chargebacks cluster in two reason codes that are nearly always refund-path failures, and neither pure automation nor pure letters wins them alone.

Physical-Goods Chargeback Defense Is a Records Problem

Physical-goods cases are won in the order log, before the chargeback arrives.

Not Every Chargeback Is Worth Fighting

Fighting every chargeback loses money on cases that should have been refunded.

Subscription Chargebacks Are Lost at Sign-up

The records that win a subscription dispute were captured at sign-up, or never.

The Chargeback Lifecycle Continues After Representment

Pre-arbitration is the issuer's second move, with fresh evidence and a network fee for the loser.

Visa's April 2026 Rule Update, Read for What Actually Matters

Two of Visa's April changes will actually reach a small merchant's desk.

A Chargeback Costs Roughly Twice What the Dashboard Says

The dashboard hides roughly half the cost of a lost chargeback.

The 7-Day Clock Inside the 30-Day Window

The acquirer's response window is a third of the network's.

The Chargeback Ratio Is the Wrong Metric

Optimising for chargeback ratio rewards refunding winners and fighting losers.

When the Customer Says the Package Never Arrived

A tracking number proves a parcel reached an address, not the right person.

Your Dispute Dashboard Was Built for Your Processor

The auto-submit default loses cases the merchant could have won.

The First Chargeback Is Won or Lost in 24 Hours

The case is decided by what the merchant captures on day one.

Rebuttal Letters Lose on Structure Before They Lose on Evidence

Every reason code wins on the same letter arc.

"Friendly Fraud" Is the Wrong Name for a Refund Problem

Most cardholders disputing their own purchases are picking the faster refund.

A Chargeback Notification Is 5 Fields and Some Boilerplate

Five fields inside the notification decide the case.

Pre-dispute Alerts Are a Tax on Bad Order Records

Paying $19 to deflect a chargeback is the cost of missing order records.

The Records That Win a Dispute Are Captured Before It Arrives

Order-time records decide the chargeback that arrives six weeks later.

VAMP Was Written for the Merchants It Hits Hardest

Visa's monitoring threshold reads as uniform but behaves as a step function.

VAMP at 12 Months: a Graduated Band, Not a Cliff

One year in, VAMP enforcement looks graduated rather than absolute.