Ask a merchant who fights their chargebacks how often they win, and the answer usually lands somewhere around "half, maybe a bit more." It feels right. They remember the disputes they won, the tracking numbers that landed, the obvious friendly fraud cases that came back in their favor. Fifty, sixty percent — that's the number most merchants will give you, and most of them believe it.
The real number is almost always lower. For merchants fighting disputes without a disciplined evidence process — the majority — actual win rates sit closer to 20 to 35 percent. The gap between the remembered number and the real one isn't small, and it isn't an accident. It comes from how merchants count, what they count, and a handful of process failures that quietly turn winnable disputes into losses. This post is about closing that gap: how win rate is actually measured, what the honest benchmarks look like, why merchants lose disputes they should win, and what the top performers do differently.
What Counts as "Winning" (and Why Merchants Miscount)
Shopify Payments gives every dispute one of a few statuses: needs response, under review, won, or lost. That's the raw material for a win rate. The problem is that almost nobody tracks it systematically. Disputes trickle in one at a time, get handled (or not) in the moment, and then disappear into the order history. Nobody is keeping a running tally, so the "win rate" a merchant quotes you is a memory, not a measurement — and memory is generous. Wins are satisfying and stick; losses are annoying and get forgotten.
The bigger error is counting the wrong things. A lot of merchants quietly fold "no chargeback was ever filed" into their sense of how well they're doing, or count a dispute they chose not to fight as something other than a loss. Neither belongs in the number. A dispute you didn't respond to is an automatic loss — the funds leave your account and it counts against your ratio exactly the same as a contested loss does. An order that never got disputed isn't a win at all; it's just a normal sale.
Define it cleanly and the inflation disappears. Your win rate is disputes won ÷ disputes contested — the share of cases you actually fought and actually won. Not disputes won over all orders. Not disputes won over disputes-you-felt-good-about. If you contested 40 disputes last quarter and the "won" status came back on 11 of them, your win rate is 27.5 percent, regardless of how the year felt. Until you're pulling that number from your dispute statuses rather than your gut, you don't have a win rate — you have an impression.
Industry Benchmarks
Before getting into numbers, an honest caveat: there is no clean public dataset for Shopify chargeback win rates. Card networks don't publish it, Shopify doesn't break it out, and most "industry average" figures you'll see floating around are either marketing from dispute vendors or surveys with small, self-selected samples. The ranges below are our best read from public industry reports, processor disclosures, and what we see across merchants — estimates, not gospel. Treat them as rough goalposts, not precise targets.
With that said, win rates vary enormously by category, and the variation tracks how strong the available evidence tends to be:
- Dropshipping — roughly 15-25%. Long shipping times, third-party fulfillment, thin tracking data, and frequent "not received" and "not as described" disputes make this the hardest category to defend. The merchant often doesn't even control the evidence.
- Subscription boxes — roughly 25-35%. Recurring billing generates a lot of "I didn't authorize this" disputes from customers who forgot they subscribed. Winnable with good consent records, but many merchants don't keep them.
- Established ecommerce — roughly 35-50%. Physical goods, domestic shipping, real tracking, and a record of customer communication. The evidence exists; whether it gets assembled and submitted well is the variable.
- Top quartile with proper evidence — roughly 60-75%. Same disputes as everyone else, but a disciplined evidence process applied to every case. This is the ceiling most merchants could reach and don't.
The point of these ranges isn't the exact figures — it's the spread. The difference between the bottom and the top isn't luck or category alone. A subscription merchant with airtight consent logs can beat the "established ecommerce" range; an established store that submits sloppy responses can fall into the dropshipping range. The evidence process moves you within and across these bands more than the category you're in.
The 5 Reasons Most Merchants Lose Disputes They Shouldn't
Most lost disputes aren't lost on the merits. The order was real, the product shipped, the customer is committing friendly fraud — and the merchant still loses. Here's where the winnable cases go wrong.
1. Missing tracking proof of delivery
For any "merchandise not received" dispute, delivery evidence is the whole ballgame, and a surprising number of merchants can't produce it cleanly. The tracking number is buried in a fulfillment email, the carrier link has expired, or the proof shows "shipped" but never "delivered." Without a tracking record that shows delivery to the order address, you're submitting a "not received" rebuttal with nothing that contradicts the claim — which is functionally an admission.
2. Generic template responses
A canned rebuttal that says "this order was valid and shipped as agreed" tells the bank reviewer nothing and signals you didn't take the case seriously. Reviewers process dozens of disputes a day; a template that doesn't reference the specific order, the specific evidence, or the specific claim reads as boilerplate and gets treated like boilerplate. The same paragraph pasted into every dispute is one of the clearest tells of a low win rate.
3. Submitting too late
Every dispute has a hard deadline — typically 7 to 21 days depending on the network — and missing it is an automatic, total loss no matter how strong your case is. Merchants who handle disputes reactively often discover them with two days left, scramble, and either submit something thin or miss the window entirely. Late and rushed is how a lot of winnable disputes quietly become losses by default.
4. Ignoring reason codes
Each dispute arrives with a reason code that tells you exactly what the bank is evaluating, and merchants routinely ignore it. They attach delivery confirmation to a fraud dispute, or fraud scores to a "not received" claim — evidence that's irrelevant to the actual question and dilutes whatever is relevant. If you don't read the code, you're answering a question the reviewer didn't ask.
5. No documentation of customer communication
One of the strongest pieces of evidence in a friendly fraud case is proof the customer never tried to resolve the issue before going to their bank — and proof requires records. Merchants who handle support across personal email, Instagram DMs, and an unlogged phone line can't produce the transcript that shows the customer had a path to resolution and skipped it. The communication happened; it just wasn't captured in a form you can submit.
What Top-Quartile Merchants Do Differently
The merchants winning 60 to 75 percent of their disputes aren't getting easier disputes. They're running a tighter process on the same cases everyone else has. The specifics matter more than the principle, so here are the actual tactics.
Submit within 48 hours. Not because the bank rewards speed, but because early submission removes the failure mode that loses the most disputes: the missed or rushed deadline. Top merchants treat a dispute notification like an order that has to ship — it gets handled the day it arrives, with the full evidence package, not on day six with whatever's at hand. Submitting early also leaves room to catch a missing piece before the window closes.
Match evidence precisely to the reason code. A fraud dispute (Visa 10.4 and its equivalents) gets AVS match, CVV match, IP geolocation, and customer order history — evidence that the cardholder placed the order. A "not received" dispute (Visa 13.1) gets tracking, delivery confirmation, and signature. They don't mix the two, and they lead with the single strongest item for that specific code rather than dumping everything in.
Include screenshots of customer messages. When a customer emailed asking about the order, or never emailed at all, that gets captured and attached. A timestamped support thread showing the customer engaging normally — or a clean record showing zero contact before the dispute — directly undercuts an "unauthorized" or "not received" claim. The top merchants keep every channel logged so this is always available.
Use AVS and CVV match records. On any fraud dispute, the Address Verification System and CVV results are the backbone of the case: they show the billing address tied to the cardholder's bank was entered correctly and whoever placed the order had the card in hand. Top merchants pull these from the order's payment details and cite them explicitly rather than leaving the reviewer to infer.
Document return policy display on the product page. For "not as described" and some friendly fraud disputes, it helps to show the customer agreed to — and could see — your return and refund terms. A screenshot of the policy as displayed on the product detail page and at checkout demonstrates the customer had a clear, non-chargeback path to resolution. It's a small piece of evidence that quietly tips the close cases.
None of this is exotic. It's the same evidence sitting in every Shopify store's order records, assembled deliberately and submitted on time. The difference between a 25 percent win rate and a 65 percent one is mostly the difference between doing this on every dispute and doing it on the ones you happen to get to.
How to Actually Close the Gap
The honest problem with the top-quartile playbook is that it's labor. Doing it on every dispute means watching for notifications, reading each reason code, pulling AVS and CVV and IP from the order, digging up tracking, finding the support thread, screenshotting the policy, writing a tailored rebuttal, and submitting before the deadline — for every case, including the ones that arrive the week you're slammed. Most merchants can do this beautifully for one dispute and not at all for the next ten. That inconsistency is exactly what drags the real win rate below the remembered one.
This is the part Paidback automates. It detects the dispute the moment it's filed, pulls the AVS match, CVV match, IP geolocation, delivery confirmation, customer history, and communication records straight from your Shopify data, matches the evidence to the specific reason code, generates a tailored response, and submits it before the deadline — on every dispute, not just the ones you have time for. It's the top-quartile process applied consistently, which is the only way the top-quartile win rate actually shows up in your numbers. If you want to stop guessing at your win rate and start raising it, install Paidback on the Shopify App Store — and if you haven't yet, the cheapest disputes are still the ones you prevent from being filed in the first place.
Paidback handles Shopify Payments chargebacks end-to-end — evidence collection, reason-code matching, response generation, and submission, all inside the Shopify admin. 15% of recovered amounts, capped at $250 per dispute. Install Paidback on the Shopify App Store.