Walmart's legal team filed a 15-page objection in a Brooklyn federal court in late 2025. The target: a proposed settlement between Visa, Mastercard, and roughly 12 million merchants that the press had framed as historic relief. Walmart called it a trivial cut that preserves the fee structure intact. The company built on per-unit cost control looked at the deal and said no.
That response tells you more than the settlement itself.
The platitude runs like this: the swipe-fee settlement will lower your processing costs. It sounds right. It is not.
What Two Reports Show
The research on this is worth reading. Two groups track swipe fees on Visa and Mastercard credit cards. The Nilson Report puts the average rate at 2.35% in 2024. CMSPI, which tracks the full cost merchants pay, puts it at 2.91%. That is a 56-basis-point gap on the same transactions, in the same year, in the same market.
The gap exists because the two reports measure different things. Nilson tracks interchange, the portion set by the card networks and paid to the issuing bank. CMSPI tracks all three layers: interchange, network assessment fees, and the processor's markup. The settlement's 10-basis-point cut applies only to interchange. And even there, the cut barely offsets a single year of increases. Interchange moved from 2.26% to 2.35% between 2023 and 2024 alone.
The settlement does not fail merchants on purpose. It fails the system they are trying to run.
The Structural Flaw
The problem is not that the cut is too small. The problem is that it addresses the one layer a merchant cannot negotiate and leaves untouched the one layer they can.
Every card transaction passes through three cost layers. Interchange goes to the bank that issued the card. Assessment fees go to Visa or Mastercard. The processor's markup goes to the company that handles your transactions. The first two are fixed. The third is not. And the third is where the money hides.
Swipesum, a payments audit firm, reports that after reviewing hundreds of merchant statements, the gap between what a business thinks it pays and what it actually pays runs 30 to 80 basis points. Sometimes more. In 2026, that gap is getting wider, not smaller.
The settlement gave back a fraction of one fixed layer while the variable layer kept growing without notice.
The Replacement Principle
The layer you can control is the one no settlement will touch for you. Start with what is true. Then build from there.
Once that is clear, three moves follow from it.
Move 1: Calculate Your Effective Rate
Pull one month's processing statement. Divide your total fees by your total volume. That single number is your effective rate. If your quoted rate is 2.1% and your effective rate is 2.9%, you have a gap of 0.8% that no one told you about. This one number tells you more than the settlement ever will, but it requires you to stop trusting the rate your rep quoted you.
Move 2: Isolate the Processor Markup
Subtract the published interchange and assessment schedules from your effective rate. What remains is markup. Some of it is clean. Some of it is not. Look for line items like PCI non-compliance charges, gateway surcharges, and "network access" fees. These are processor markup dressed as pass-through costs.
Move 3: Bid the Account
Take your effective rate and the line-item breakdown to a second processor. Request interchange-plus pricing, where the processor shows the base cost and adds a flat margin on top. On $5 million in annual volume, a gap of 0.7% costs $35,000 per year. That is not a rounding error. That is a hire.
What the System Shows
Running this for 90 days does something the settlement never did:
You see the true cost of each transaction type, not the blended average your statement shows. You see which line items are fixed and which are inflated. You see how your effective rate shifts month to month, which tells you whether your processor is adjusting fees without telling you. And you have a number to bring to a negotiation, not a hope that a court ruling will do the work for you.
The Feedback Loop
At the end of 90 days, ask three things.
→ What moved the number? Name the specific line items or rate changes that shifted your effective rate.
→ What looked like progress but left no trace? A new contract that promised savings but did not change the effective rate is not a win. It is a repackaging.
→ What friction showed up more than once? If a fee keeps appearing under different names, the processor is not making a mistake. They are making a margin.
That is the difference between a settlement that sounds right and a system that proves itself.
Everything is coming into focus.
Join beehiiv live on July 16th at 1PM ET for a first look at the future of audience-led business.
This isn’t just another feature launch (though there will be plenty of those). It’s a look at a more connected future for creators and brands that are tired of juggling disconnected tools, platforms, and data.
If you care about building an audience online, this is worth your time.
Where You Stand
Walmart did not wait for a settlement to fix its processing costs. It built the system to see the gap, then used what it found. The version of that system available to a small operator fits on one page and starts with one month's statement.
Total swipe fees have more than doubled in the past decade. The settlement trims a sliver of one layer for five years. The layer that costs you the most is the layer you can read, measure, and renegotiate on your own.
Your processing statement is not a receipt. It is a negotiation you have not started yet.
