When Your Only Processor Goes Down, So Does Your Store

Compaytence Brief · August 10, 2026

PayPal's Braintree GraphQL API went down recently for roughly two hours. Merchants running payments through it saw HTTP 500 errors across checkout, withdrawals, and Venmo transfers, all at once, with no fallback to route around it.

For those two hours, customers weren't thinking about PayPal. They were staring at a broken checkout on your store, a card that should have worked but didn't, or a cart that wouldn't submit. Every one of those attempts is revenue you don't get back, and a real share of those customers won't come back later to try again. An outage doesn't just cost you the sale in the moment, it costs you the trust that got someone to checkout in the first place, at your expense, not your processor's.

Two Ways to Go Dark

When your entire payment stack runs through one processor, their problem becomes your problem, and it happens in two different ways.

The first is the kind you can't predict: an infrastructure failure like the one above. There's no partial degradation, checkout either works or it doesn't, and there's no second rail to route transactions to while you wait it out.

The second is worse, and it isn't an accident. Major processors like Stripe, PayPal, and Shopify Payments approve merchants instantly and underwrite after the fact, which means a volume spike or a chargeback cluster can trigger a hold with no warning at all. Rolling reserves lock up 10 to 25% of revenue for 90 to 180 days during that review, and if the account gets terminated rather than paused, MATCH listing can block approvals with other processors for up to five years.

Where an outage costs you hours, a shutdown can cost you months. To a customer trying to check out, both look exactly the same: your store isn't working.

Enterprise merchants already know this. In an ACI Worldwide survey of more than 100 Tier 1 merchants doing $500M+ in annual revenue, 97% run multiple acquirers. Of those, 40% saw roughly a 1% lift in acceptance rates and two-thirds cut processing costs by at least 2%.

The Bigger Problem: One Processor Isn't Built for Everything You Sell

Redundancy solves the outage and the freeze. It doesn't solve the second problem, which is that most single-processor setups are also the wrong processor for at least part of the business.

Vertical fit. Standard aggregators aren't built to underwrite every business model. CBD, nutraceuticals, subscriptions, and other categories with unpredictable refund patterns often get capped, flagged, or declined outright by generalist processors, not because the business is doing anything wrong, but because the risk profile doesn't match what the aggregator underwrites for. Specialized high-risk processors charge more for it (fee ranges around 2.3 to 3.4% for mid-risk and 2.7 to 4.3% for high-risk, per PaymentCloud's published rates), but they actually support the model instead of shutting it down at the first refund spike.

Volume fit. A processor that's fine at $20k a month becomes a liability at $500k a month. Aggregators run automated velocity triggers that were never designed for fast scaling, which is exactly when a sudden hold does the most damage. Enterprise-grade acquiring relationships exist for a reason, they're built for volume an aggregator's risk models weren't tuned for.

Regional fit. International cards aren't the default payment method everywhere you might sell. In Latin America, a $215B+ eCommerce market growing 12.2% year over year, cards account for roughly 42% of transactions while alternative payment methods, wallets, QR-based rails, cash-to-digital options, are expanding fast. As ACI Worldwide puts it, no single checkout strategy converts across the region. A processor built for U.S. card volume simply won't clear the local rails your international customers actually use.

The Actual Takeaway

A payment stack fails for a specific set of reasons: a shutdown with no warning, a reserve that locks up cash flow, an authorization rate that quietly caps how much revenue actually clears, or a currency and payment method gap that loses the sale before a transaction is even attempted. Vertical, volume, and region are what create that exposure, but the exposure itself is what to measure.

Start narrow: pull your current reserve terms, your authorization rate over the last 90 days, and a list of the currencies and payment methods you don't support. Whichever one is furthest off is where the setup needs to change first.

How Compaytence Fits

This is the exact problem Payment Setup is built to solve. Compaytence designs custom payment stacks for international eCommerce merchants, built around your specific vertical, volume, and regions instead of routed through whatever generalist aggregator approved you fastest.

That means no default exposure to sudden shutdowns, standard rolling reserves, or a single point of failure when one processor goes down. It also means access to the payment methods and currencies your customers actually use at checkout, and higher authorization rates from routing transactions through tier-one providers matched to your business instead of a one-size-fits-all risk model.

If you want to see where your current setup is exposed, that's the first thing we map out in a Payment Setup audit, run against a network of 30+ top-tier providers.

Book a Call →