Chargeback Reason Codes Explained: The Complete 2026 Guide

Every chargeback arrives with a code attached, and that code is not bureaucratic noise. It is the single most important piece of information in the dispute, because it tells you precisely what the cardholder is claiming and, by extension, exactly what you must prove to win.

Merchants lose winnable disputes every day by submitting the wrong evidence for the code they were given. This guide is the complete reference: what reason codes are, the four universal categories every code belongs to, how each network numbers them, a master table of the most common codes across Visa, Mastercard, Amex, and Discover, and how to turn any code into a winning rebuttal.

What a chargeback reason code actually is

A chargeback reason code is a short alphanumeric label the card network assigns to a dispute to classify why the cardholder is challenging the charge. When a customer contacts their bank to dispute a transaction, the bank selects a code that best matches the complaint. That code travels with the dispute all the way to you.

The code does two things:

1. It states the claim. “The cardholder says they never received the item.” “The cardholder says they did not authorize this.” “The cardholder says they were charged twice.”
2. It sets the evidence bar. Each code has a defined set of evidence the networks will accept as a rebuttal. Delivery confirmation wins a not-received dispute. It does nothing for a fraud dispute.

Read the code first. Everything you do afterward — what evidence you gather, how you write the rebuttal, whether you fight at all — flows from it.

The four universal categories

Every reason code across every network maps to one of four underlying categories. Networks number them differently, but the logic is identical. Once you can place a code in its category, you know the shape of the dispute even before you learn the network’s specifics.

The overwhelming majority of ecommerce chargebacks fall into just two of these: fraud and consumer disputes. Authorization and processing-error disputes are real but comparatively rare for card-not-present merchants.

A note on the fraud category: much of it is not true criminal fraud at all. Industry data shows friendly fraud — a real cardholder disputing a charge they actually made — drives well over half of all chargebacks and is the fastest-growing type. Those disputes arrive dressed as fraud codes, which is precisely why matching evidence to the code (and disproving the fraud claim) matters so much. For more, see our breakdown of friendly fraud.

How each network numbers its codes

The four categories are universal. The numbering is not. Each network has its own scheme, and you need to recognize all four because a single processor account can present disputes from every network.

Visa

Visa uses a two-part decimal format organized by category. The number before the dot is the category:

• 10.x — Fraud (e.g., 10.4 Fraud Card-Absent Environment)
• 11.x — Authorization
• 12.x — Processing Errors
• 13.x — Consumer Disputes

This is the cleanest system of the four because the category is baked right into the code. See the full list in our Visa chargeback reason codes guide.

Mastercard

Mastercard uses four-digit codes (all starting with 48). The categories are not encoded in the number, so you have to know them:

• 4837, 4863, 4870, 4871 — Fraud and chip-liability
• 4808 — Authorization
• 4834 — Processing/point-of-interaction errors
• 4853, 4855 — Consumer disputes

Notably, Mastercard consolidated many older consumer-dispute codes into a single broad code, 4853 Cardholder Dispute, covering not-received, not-as-described, defective, and credit-not-processed. Full detail in our Mastercard chargeback reason codes guide.

American Express

Amex, as both network and issuer, uses letter-plus-number codes grouped by first letter:

• F-codes (F10, F24, F29, F30) — Fraud
• A-codes (A01, A02, A08) — Authorization
• P-codes (P01, P05, P08) — Processing errors
• C-codes (C02, C08, C18, C31) — Consumer disputes (e.g., C08 goods/services not received)

Discover

Discover, also its own network and issuer, uses two-letter codes:

• UA — Fraud (e.g., UA01 card-present fraud, no authorization)
• AA — Cardholder does not recognize / does not agree
• RG — Non-receipt of goods or services
• RM — Quality: goods or services not as described
• RN — Credit not processed

Master reference: the most common codes

This table covers the codes card-not-present merchants actually encounter, grouped by category, with what the cardholder is claiming and the evidence that wins. Use it as your first stop when a dispute arrives.

Codes evolve, and networks periodically renumber or consolidate them. Treat this table as the working set for 2026, and always confirm the code as your processor reports it.

Turning a code into a winning rebuttal

A reason code is a question. Your rebuttal is the answer, and the answer has to be specific to the question asked. Here is the workflow that wins:

1. Identify the code and its category. Is this fraud, authorization, a processing error, or a consumer dispute? The category tells you the general shape; the exact code tightens it.
2. Restate the claim in plain terms. “The cardholder says the item never arrived.” Writing it out prevents you from arguing the wrong point.
3. Gather the evidence that maps to the code, and only that evidence. For a not-received dispute (13.1 / C08 / RG), that is delivery confirmation and tracking. For a fraud dispute (10.4 / 4837), it is AVS/CVV, IP/device, delivery to the cardholder, and — critically — prior undisputed transactions from the same customer.
4. Write a tight rebuttal that ties each piece of evidence to the claim. Reviewers spend seconds per case. Make the contradiction obvious: here is the code, here is the claim, here is the proof it is wrong.
5. Submit before the processor deadline. The best evidence loses if it arrives late.

The single most common mistake is a mismatch: submitting delivery tracking to fight a fraud claim, or AVS data to fight a not-received claim. The evidence might be true, but if it does not answer the code’s specific claim, it does not win. For the full method, see how to win a chargeback dispute.

Where the reason code fits in the bigger picture

Fraud codes deserve special attention because they are the most common and the most winnable when handled correctly. Visa’s flagship ecommerce fraud code, 10.4 (Fraud, Card-Absent Environment), now has a dedicated winning path through Compelling Evidence 3.0 (CE 3.0), which lets you defeat a fraud claim by showing a history of prior undisputed transactions from the same customer sharing data points like IP address, device ID, and delivery address. If you sell online, this is likely the code you will see most, and it is worth understanding in depth — see our guide to Visa reason code 10.4.

The bottom line

The reason code is the dispute. It states the cardholder’s claim and sets the evidence bar, and matching your rebuttal to it is the difference between winning and losing a case you had every right to win. Learn the four categories, recognize each network’s numbering, and build every rebuttal to answer the specific code you were given.

This is also exactly the kind of rules-based, evidence-matching work that automation does better than a human racing a deadline. DisputeDash reads the reason code the moment your processor reports it, gathers the code-specific evidence automatically, and builds the matched rebuttal — so the right proof answers the right claim, every time.