Why a dispute costs $229 on a $129 pair of shoes · AR

AR

An engineer (and manager) at Brex focusing on Fintech. <3 automated tests and pair programming. Find me at @allcentury.

Home
About
Payments Index
GitHub
Currently v2.1.0

© 2026. All rights reserved.

Why a dispute costs $229 on a $129 pair of shoes
02 Sep 2026
Part 2 of 2 in Payments Disputes
Hey, if you’re new here, I’m Anthony Ross. I’ve spent the last 10+ years working in fintech (Brex + Braintree/Venmo). In my last post I walked through what happens the moment you tap dispute, the five parties, the fraud vs. everything-else fork, how much time everyone actually has, etc. This one answers the question that actually matters to a merchant: once the dispute lands, who pays for it? or rather, who pays and who gets paid and where did all my money go…
The answer to who loses isn’t “the person who did something wrong.” It’s usually “whoever has the weaker evidence or tech” which is a very different thing, and it’s why disputes are so expensive.
Proof of Delivery
A “I never received the goods” dispute is very common, especially for e-commerce companies and the dispute is extra costly because there’s no liability-shift rule for it and no clean policy answer, just evidence or the lack of it. If you can’t prove delivery, you lose, full stop, regardless of whether the customer is telling the truth or not.
Here’s what that actually costs. Say you sold a pair of shoes for $129, shipping costs you $11, and your card processing fee was 3%, $3.87. A dispute comes in for the full $129, and you don’t have delivery evidence, so you lose. Here’s the damage:
lost sale, reversed by the chargeback $129.00
shipping, already paid, gone $11.00
card fee, not refunded on a chargeback $3.87
cost of the shoes, gone $70.00
flat dispute fee, charged either way $15.00
------------------------------------------------
total loss $228.87

$228.87 lost on a $129 sale, 1.77x the original transaction, and that’s a mild case. Lower margin goods hurt worse and with enough of these, your acquirer will put you into a monitoring program that can raise your acquiring rate + disptue fees. The chargeback fee alone is worth calling out, it’s non-refundable even if the merchant fights the dispute and wins later. Winning gets you your $129 back. It does not get you your $15 back, and on some smaller transactions, $15 might be greater than your actual margin.
Card-present fraud: the EMV liability shift
Contrast that with card-present fraud, where the networks actually did engineer a more obvious answer. If you’ve ever wondered why every card terminal on earth suddenly wanted you to insert your chip instead of swipe, this is why. EMV stands for Europay, Mastercard, and Visa, the three companies that created the chip standard, and the rule, in plain terms, is that liability falls on whichever side is using the weaker technology.

When a Merchant doesn’t support a chip reader, or has a chip reader but processes it as a swipe anyway, and the transaction turns out fraudulent, the merchant eats it - everytime.
When a Merchant has a chip-enabled terminal, but the card itself is an old mag-stripe-only card with no chip to insert, so it gets swiped instead, the issuer eats it (usually).
Both sides are chip-compliant and the transaction still turns out to be counterfeit fraud (the chip got cloned some other way), the issuer is typically still on the hook. Neither party did anything wrong, so the loss falls back to whoever’s supposed to be backstopping fraud in the first place.

That’s a genuinely elegant piece of policy design, actually, it doesn’t try to figure out who’s at fault, it just makes upgrading your security the economically rational move for everyone. Once EMV adoption crossed a threshold, this stopped being a live problem for most merchants. The schemes (Visa, Mastercard, etc) know how to incentivize fraud measures, by pushing the cost back to the weakest link in the payments chain.
Why non-delivery is the one that stays expensive
The EMV shift is a policy lever, upgrade your tech and the risk moves off you. Non-delivery disputes don’t have that lever. The only thing standing between you and this $228.87 problem is whether you can produce a delivery confirmation, a signature, tracking that actually shows it arrived, something. No evidence means no defense and it doesn’t matter that you actually shipped the shoes.
That’s the setup for the next post. Merchants aren’t defenseless here, they can fight back with evidence, and what actually counts as evidence, what doesn’t, and why most merchants don’t bother even when they have a winning case, is where representment comes in.

Related Posts

What actually happens when you tap "dispute this charge"
30 Aug 2026

Making tmux.expose agent-aware
27 Aug 2026

Making tmux.expose match my setup
26 Aug 2026

A dispute incurs significant financial costs for merchants, which is often determined not by who committed the initial wrong, but by who possesses the weaker evidence or technology. This dynamic explains the high expense associated with payment disputes. For instance, a dispute concerning non-delivery of goods, common in e-commerce, carries a substantial cost because there is no universally applied liability-shift rule or clear policy framework, meaning the loss is often determined solely by the presence or absence of verifiable evidence. Using an example of selling shoes for one hundred twenty-nine dollars highlights this expense: a dispute can result in a total loss approximating two hundred twenty-eight dollars and eighty-seven cents, calculated by factoring in the lost sale, already paid shipping, card fees, the cost of the goods, and a flat dispute fee. This loss is magnified for lower margin items, and repeated instances can lead acquirers to impose monitoring programs that increase acquiring rates and dispute fees.

In contrast to non-delivery disputes, card-present fraud exhibits a more structured resolution mechanism known as the EMV liability shift. EMV, which stands for Europay, Mastercard, and Visa, established a rule where liability for fraud falls upon the party utilizing the weaker security technology. This mechanism incentivizes all parties to upgrade their security protocols. For example, if a merchant accepts a transaction via swipe despite having a chip reader, or if a cardholder uses a mag-stripe-only card when a chip is enabled, the liability shifts to the party with the weaker mechanism. This policy design is considered elegant because it does not focus on assigning fault but rather makes security upgrades the economically rational choice for everyone involved in the payments chain. As the EMV adoption threshold was crossed, this liability-shifting approach effectively managed fraud risks for most merchants.

The fundamental difference in how these disputes are handled stems from the presence of an external policy lever. Non-delivery disputes lack this structural mechanism, meaning the only way for a merchant to defend against the financial loss is by successfully producing concrete evidence, such as delivery confirmations or tracking information. The ability to establish this evidence determines the outcome, irrespective of whether the shipment was actually delivered. Consequently, the capacity to produce verifiable proof is the key factor in mitigating the extensive losses associated with non-delivery claims. Furthermore, the text suggests that the determination of what constitutes sufficient evidence and the resulting defensive strategy are deeply tied to the concept of representment.