Most merchants find out about a card testing attack the same way: a flood of declined transactions at 3:00 in the morning, a gateway bill full of authorization fees, and a phone call from the processor asking what happened. Here at CyoGate we see these attacks every week, and they almost always hit merchants who had no idea their checkout page was an easy target.

A card testing attack (the card networks call it "enumeration") is when a fraudster uses a bot to run stolen or computer-generated card numbers through your payment form to find out which ones work. The cards that approve get sold or used somewhere else. You get stuck with the fees, the chargebacks, and the damage to your merchant account. The good news is that card testing is one of the most preventable kinds of fraud there is, once you know what to look for.

What a Card Testing Attack Looks Like

Card testers want volume and speed, and that leaves a recognizable pattern in your transaction report. In our experience, the warning signs show up in this order:

  • A sudden spike in declines. Your normal decline rate might be 5% to 10%. During an attack it can jump past 80% because most of the card numbers are invalid.
  • Lots of small transactions. Testers like $0.50, $1.00, or your cheapest product, because small charges are less likely to get flagged by the cardholder's bank.
  • Many different cards from one source. Dozens or hundreds of card numbers coming from the same IP address, device, or email domain in a short window.
  • Odd timing. Activity in the middle of the night, or in bursts of transactions seconds apart, which no human customer does.
  • Nonsense billing details. Random names, mismatched ZIP codes, and addresses that fail AVS (the Address Verification Service check that compares the billing address to the one on file with the card issuer).

Donation forms, low-priced digital products, and any page that runs a $0 or $1 card verification are favorite targets. If your checkout lets someone submit a card without creating an account or solving any kind of challenge, assume testers will find it eventually.

Why Card Testing Is a Bigger Problem Than It Used to Be

For years, card testing was mostly an annoyance: wasted authorization fees and a messy report. That changed when Visa rolled out the Visa Acquirer Monitoring Program, known as VAMP. We covered the full program in our post on what changed with Visa VAMP, but the short version is that enumeration now counts against you directly.

VAMP tracks an enumeration ratio, which is the share of your card-not-present transactions that Visa flags as card testing. A merchant at or above 20% is classified as excessive. VAMP also tracks a combined fraud and dispute ratio, and Visa lowered the excessive merchant threshold to 150 basis points (1.5%) in the U.S. on April 1, 2026. A single bad weekend of card testing can push a small merchant over both lines, because the approved test charges often come back later as fraud reports and chargebacks.

Then there's your processor. Acquiring banks pay attention to sudden decline spikes because Visa now holds them responsible too. We have seen processors freeze or terminate accounts after one serious attack, and a termination for fraud can land a business on the MATCH list, which follows you for five years. For a high risk merchant who already worked hard to get approved, that is a disaster that was completely avoidable.

The Real Cost of a Card Testing Attack

Merchants usually underestimate what a card testing attack costs because the losses show up in several places at once:

  • Authorization fees. You pay a per-transaction fee on every attempt, approved or declined. Ten thousand attempts adds up fast.
  • Chargebacks and refunds. The test charges that approve end up as fraud chargebacks when the real cardholders spot them, each one carrying a chargeback fee and a mark on your ratio.
  • Monitoring program exposure. Crossing VAMP thresholds can bring network fines that your acquirer typically passes along to you.
  • Account risk. Higher reserves, held funds, or outright termination.
  • Lost sales. If your processor shuts off card-not-present processing during an investigation, your real customers can't buy either.

How to Stop Card Testing: Gateway Settings That Work

No single tool stops every bot. What works is layering several simple controls so that testing your checkout becomes slow, expensive, and not worth the effort. Here is the setup we recommend to CyoGate merchants.

1. Set velocity limits

Velocity limits cap how many transactions can come from one IP address, one card number, or one email address within a set time. A real customer rarely tries more than two or three cards in an hour. A bot tries hundreds. Our iSpy fraud prevention rules let you set these limits at the gateway level, so bad attempts get blocked before they ever reach the card networks. Our guide to rule-based fraud screening with iSpy walks through the most useful rules by merchant type.

2. Require CVV and AVS

Make the card security code (CVV, the three or four digits on the card) a required field and decline transactions that fail the CVV check. Testers working from generated numbers usually don't have it. Add AVS rules that reject clear mismatches, while allowing partial matches for international customers if you sell overseas.

3. Add a challenge to your payment form

A CAPTCHA or invisible bot check on the checkout and donation pages stops most automated scripts cold. Put it in front of the payment submission, not just the login. If a bot can reach your payment form without passing a challenge, the form is the attack surface.

4. Set a minimum transaction amount

If nothing you sell costs less than $10, there is no reason to accept a $1 charge. A minimum amount rule shuts down the micro-charges testers depend on. Nonprofits that accept donations of any size should set a sensible floor, such as $5, since donation forms are among the most heavily tested pages online.

5. Turn on 3D Secure

3D Secure authentication asks the card issuer to verify the cardholder, usually invisibly and sometimes with a one-time code. Stolen card numbers tend to fail that step, and authenticated transactions shift fraud chargeback liability to the issuer. Our 3D Secure 2.0 guide covers how to roll it out without hurting conversion.

6. Protect your tokenization keys

If you use hosted tokenization such as Collect.js, your card data stays off your server, which is great for PCI compliance. But tokenization alone does not stop a bot from using your form. Watch for unusual token creation volume and rotate your public tokenization key if you suspect it is being abused from outside your site.

7. Block repeat offenders

Once you identify the IP ranges, BIN ranges (the first six to eight digits of a card number that identify the issuing bank), or countries an attack is coming from, add block rules. Review them every few weeks so you don't end up turning away good customers long after the attack has ended.

What to Do If You're Under Attack Right Now

If you are seeing a decline spike as you read this, move quickly:

  1. Temporarily disable the targeted form or product, or switch on a strict velocity limit across the board.
  2. Call your processor and tell them what is happening. Processors are far more forgiving when the merchant reports the attack first.
  3. Pull the transaction report and void any approved test charges that haven't settled yet. Refund any that have, before they turn into chargebacks.
  4. Add the IP addresses and card ranges involved to your block list.
  5. Put the layered controls above in place before you turn the form back on.

Refunding approved test charges early matters more than most merchants realize. A refund issued before the cardholder disputes the charge keeps it from becoming a chargeback, which protects your ratio. Our post on the warning signs that your chargeback ratio is creeping up explains why those numbers matter so much.

Card Testing and High Risk Merchant Accounts

High risk merchants feel card testing the hardest. Their accounts often come with tighter chargeback thresholds, rolling reserves, and less patience from the acquiring bank. We have found that the merchants who keep their accounts long term treat fraud screening as part of running the business, not something they bolt on after a problem. Our broader guide to fraud prevention for high risk industries goes deeper on industry-specific patterns.

Every CyoGate high risk merchant account runs on a gateway with iSpy fraud rules, 3D Secure, and hosted tokenization built in, and our team will help you configure them for your business. If your current processor leaves you on your own when bots show up, apply with CyoGate and tell us about your checkout and traffic. We'll help you lock it down.