I can tell you the exact moment a pricing conversation changes. An operator asks what a payment rail actually costs, and instead of a number, they get a form. Then a follow-up call. Then a custom quote, days later, built around whatever the sales team thinks that account can bear. By the time an answer arrives, the conversation has stopped being about evaluating infrastructure and started being about negotiating access to a number that should have been public from the start.

Mastercard and FXC Intelligence put a name on this in June: the uncertainty tax. Research across Mexico, Brazil, and Colombia found that small and mid-sized businesses often can't get straight answers to three basic questions before sending a payment: when it will arrive, how much will land, and what gets deducted along the way. About one in nine cross-border payments needs investigation or repair, costing up to three and a half hours of staff time to fix. Nine out of ten businesses interviewed said they'd consider switching providers over exactly this, with up to 70 percent of their payment volume at risk of moving.

That's the piece I think most of the industry still gets wrong: pricing opacity manufactures the exact uncertainty that makes clients leave, while everyone treats it as a way to protect margin. Every hour a business spends chasing down what a payment actually cost them is an hour they remember the next time they choose a provider.

Part of the excuse is real. Mexico settles through SPEI, Brazil through PIX, Colombia through PSE, and Argentina layers its own capital controls on top of all of it, so a single price sheet genuinely doesn't work across the region. But that complexity is a reason to publish more rate detail by corridor, not a reason to publish none of it. Businesses moving real volume have already done the math on what things should cost. What they're missing is a real number, not a lower one.

The three and a half hours of staff time Mastercard's research points to isn't abstract to me. That's a finance team member who could be closing a books cycle instead of calling a support line to find out why a payment landed short. At Yellow Card, our API pricing for the region is published and documented, real rates in Mexican pesos, Argentine pesos, Colombian pesos, and more, readable before anyone writes a line of code. That doesn't solve the whole problem, but it proves publishing rates in a region this complex is possible. Plenty of people in this industry still treat that as aspirational. We're already doing it.

To be sure, critics inside this industry will argue custom pricing exists because volume, risk, and history genuinely differ client to client, and that publishing a flat rate ignores that. That's fair. The fix is a rate table that varies by corridor and volume tier, published instead of negotiated case by case. Pricing has always varied. What's missing is that most of it never got written down anywhere a client could see before they had to ask.

The operator I described at the start eventually got a number, but only after a form, a follow-up, and a wait he shouldn't have had to sit through. That wait is a choice the industry keeps making. If you're building payment infrastructure for this region, or buying it, ask the plainest question there is: how long does it take to get a real number? If the answer is days, that's the product telling you something about itself.