How many chargebacks your account can take before they cut off your card payments
By the end, you'll know that your chargeback ratio counts every case that's opened, whether you win it, lose it, or let it go, and that you protect your card processing by preventing new cases from opening, not by giving up on the ones you can win.
By Yodai · Published September 6, 2026 · Updated October 13, 2026
You fought the chargeback. You have proof of delivery, the WhatsApp message where the customer confirms the appointment, the signed order. You're going to win.
And yet your payment processor writes to you: account under review, risk of losing card payment processing. How can this happen if the case is already won?
It happens because the bank and the card networks don't count cases won or lost. They count cases opened. And that number, divided by your sales for the month, is what decides whether you keep taking card payments or not.
The number that counts isn't the case, it's the ratio
You already know, from the last lesson, the dollar amount at which it's worth fighting a single chargeback. That math still works for deciding whether to fight each case. But there's a number running behind every case, win or lose, and that number doesn't forgive: your chargeback ratio.
The ratio is simple: chargebacks opened that month, divided by the number of sales charged to a card. If you had 400 sales and 3 chargebacks in March, your ratio is 0.75%. It doesn't matter if you won two of those three. All three count, because they count on the day they're opened, not the day they're resolved. That's straight from Stripe's documentation on monitoring programs: those programs don't factor in the outcome of the dispute, and each case gets assigned to the month the network receives it, no matter when the charge happened.
That's the trap. You think of each chargeback as a case you win or lose. Card networks see it differently: every filing is a signal that something about how you sell isn't convincing the customer enough to keep them from calling the bank before calling you. It doesn't matter how the case ends, the signal is already on record. Stripe's own page puts it plainly: networks care more about how well you prevent disputes than whether you win them.
The ratio is calculated on cases opened, not on cases resolved in your favor
The thresholds that actually apply
Your bank doesn't set these numbers on a whim. Visa and Mastercard set them, and every payment processor adds its own safety margin on top, because the processor gets measured too: according to the official fact sheet for Visa's monitoring program (VAMP), an acquirer gets flagged once the ratio across its entire portfolio of businesses hits 0.5%, and hits the excessive tier at 0.7%. If you land in a monitoring program, the bill and the risk land on your processor too. Visa reviews last month's numbers every month, and when it counts, it adds in the fraud alerts that come from the customer's bank, even ones that never turn into a chargeback. Mastercard divides this month's chargebacks by last month's sales, so a slow sales month makes the same number of cases weigh more.
Notice the order here. Your processor doesn't wait for the grace months that, according to Solidgate (a payment processor), Visa allows, or for you to hit the 1,500 cases a month that mark the excessive tier. And don't let that number confuse you: Visa puts you under watch (the non-compliant tier) starting at just 5 cases and a 0.5% ratio, and can charge you fees, so even a small business can land in its program. According to Stripe's page on measuring disputes, the industry considers dispute activity above 0.75% excessive, and a sharp spike or an upward trend can land you in a monitoring program before you ever reach that number. Your processor acts on its own judgment and its own timeline: Stripe, for example, says on that same page that it reaches out to you if it sees your disputes climbing.
This changes the question you need to ask. It's not just whether you'll win this chargeback. It's how many chargebacks you've gotten this month, and what ratio that gives you against what you've sold. That second question doesn't show up in any warning email until it's already too late to fix it that month.
Fighting it or letting it go doesn't move the ratio
- You fight every chargeback above the threshold that makes it worth your time, as you saw in the last lesson
- Each case is judged on its own, weighing what the evidence costs against what you'd lose
- You count the chargebacks you've gotten this month and divide them by your card sales
- If you're getting close to your processor's limit, the urgent thing is to stop new ones from opening: respond faster to anyone with a complaint and refund them if that's warranted, before they go to the bank
- You look for which sale or which reason keeps repeating, so you can cut it off at the root
- You look at the month as a whole before you look at the single case
Which one I'd pick: Both, because they decide different things. A chargeback you've already gotten counts toward the ratio whether you fight it or not, so you fight it if the amount is worth it, even when you're tight on ratio: letting it go just costs you the money. The ratio decides something else: how much urgency you put into preventing the next one.
Keep in mind that getting close doesn't mean waiting for your processor's exact number. If your processor's bar is 0.75%, start watching closely once you hit 0.3% or 0.4%. That margin is what gives you time to react before your bank makes the decision for you, instead of the other way around.
So how do you check this without keeping a separate spreadsheet? Add up the chargebacks you've gotten this month, using the email alerts your processor sends you, and divide them by the card sales you took in during that same period. It's a two-number calculation, not an audit. Many processors already show it to you: Stripe, for example, displays your account's dispute activity in the analytics section of its dashboard. Do it once a week, not once a month, because the month closes before you have time to react if you wait until the thirtieth.
The costly mistake
This doesn't mean you stop fighting the chargebacks that are worth fighting: fighting them doesn't raise the ratio, and letting them go doesn't lower it. It means you're looking at two different things: the dollar amount, to decide whether to fight each case that comes in, and how many you've had this month, to know how much room you have left before your processor's warning. When that room shrinks, the urgent thing is no longer winning cases, it's stopping new ones from opening. Stripe's guide to preventing fraud puts it this way: if you respond to problems quickly and refund the money or replace the order without delay, the customer is less likely to bother disputing the charge. Refunding after the chargeback has already been opened doesn't help here: monitoring programs don't count refunds when they tally disputes. The next question, when the same reason keeps repeating month after month, isn't how hard to fight, it's what you change about how you sell.
- You know the ratio counts chargebacks opened, not cases won or lost
- You know Visa's thresholds, which start at 5 cases and 0.5%, Mastercard's, and your processor's, which acts on its own judgment
- You know that fighting or letting go of a case doesn't move the ratio: you fight each case based on the dollar amount, and you protect the ratio by preventing new cases from opening
- You check your ratio once a week, not once a month
Test yourself
Answer before opening each one. If one gives you trouble, go back to the section that explains it.
1. You've had two chargebacks this month out of 300 card sales, a 0.67% ratio, and a third one comes in for a $180 order that you're completely sure you'll win. Do you let it go to protect the ratio?
No. That third one already counts, whether you fight it or not: three out of 300 sales puts you at 1%, above the 0.75% the industry considers excessive. Letting it go just costs you the $180, and refunding it now doesn't take it out of the count either. Fight it, because with a case this clear the amount is worth it, and focus on the urgent part: keeping a fourth one from opening. Respond today to any customer with a pending complaint and, if it's warranted, refund them before they call the bank.
2. Your ratio this month is 0.1%, and you get a chargeback for a $220 service with all your evidence in order. What do you decide?
You fight it, same as the amount test from the last lesson called for. What decides is the cost of fighting versus what you'd recover, and the math works the same even with a high ratio: the case already counts against you the moment it's opened, whether you fight it or not. What a low ratio buys you is peace of mind: no rush to change anything else.
3. Your payment processor watches your account starting at a 0.75% ratio. Visa puts you on watch (non-compliant level) starting at 5 cases and a 0.5% ratio, and can charge you fees; the excessive level, with a guaranteed fine, is 1.5% and 1,500 cases. Which threshold do you pay attention to?
Both. With 5 cases in a month and a 0.5% ratio, you're already in Visa's non-compliant level, which can charge you fees, even if you're nowhere near the 1,500 cases of the excessive level. And your processor can act at 0.75%, or sooner if it sees a sharp jump. Focusing only on Visa's excessive level leaves you exposed on both fronts.
Winning cases doesn't lower the ratio, and letting them go doesn't either. What lowers it is fewer cases getting opened in the first place. If the same reason keeps coming up next quarter, the problem isn't how you're fighting the chargeback anymore, it's how you're selling.
Sources
- Visa Acquirer Monitoring Program Overview. Visa, 2025.
- Dispute and fraud card monitoring programs. Stripe Documentation, accessed October 2026.
- Measuring disputes. Stripe Documentation, accessed October 2026.
- MasterCard Excessive Chargeback - Merchant Program Guide. J.P. Morgan Chase Merchant Services, 2019.
- Visa monitoring programs (VAMP). Solidgate Docs, accessed October 2026.
- Best practices for preventing fraud. Stripe Documentation, accessed October 2026.