When credit card processing fails, the first thing to determine is whether the problem is on your end or at the processor. That distinction changes what you can do in the next ten minutes, and it changes how much this is going to cost you.
Key takeaways:
- Consumers abandon a purchase after ~7 minutes; the average outage lasts 2 hours, meaning you will lose most customers before the problem is fixed
- $44.4 billion in U.S. retail and hospitality sales is at risk annually from payment outages
- 71% of enterprise merchants run the majority of their volume through a single processor; only 32% have automated backup routing
- Manual failover is too slow: the revenue loss curve peaks between minutes 8 and 13, before most teams can even respond
- Automated payment failover routes transactions to a backup processor in milliseconds, with no action required from your team
Already know you need automated failover and want the setup details, not the background? See how Payment Gateway Failover works.
What’s happening when credit card processing goes down
Payment processing failures split into two types: problems on your side (hardware, network connection, software configuration) and problems at the processor or card network level.
If the issue is local, some steps can help: restart the terminal, check the network, try a different device. If the issue is at the processor, none of that changes the outcome. Your authorization request reaches the processor, the processor can’t complete it, and the transaction fails regardless of what you do on your end.
Both types look identical to your customer: a declined or stalled payment. The difference is in how long it takes to resolve and what options you have while you wait.
What causes card processing outages
Most credit card processing outages trace back to one of four causes:
- Internet connectivity and network failures: the most common for individual merchants. If the network connection drops, your POS terminal or checkout form can’t reach the payment gateway.
- Processor and gateway system failures: less frequent but affect many businesses at once. In August 2021, a TSYS outage disrupted hundreds of thousands of businesses nationwide, including 200,000 clients of Shift4 Payments, for approximately three hours (Digital Transactions). In May 2025, a Fiserv outage took more than 12 hours to resolve, disrupting Zelle, ACH payments, bill pay, and direct deposits across Bank of America, Capital One, and Navy Federal Credit Union (American Banker).
- Hardware failures: affect individual POS terminals and card readers, typically at a single location rather than across a whole network.
- Card network outages: rare but high-impact at the Visa or Mastercard level. When Visa’s European network failed in 2018, it produced 5 million failed transactions in 10 hours (Forbes Technology Council, November 2024).
Nearly two-thirds of payment disruptions occur during peak business hours (Dynatrace, FreedomPay, and Retail Economics, January 2026). An outage at 2 PM on a Friday costs more than the same outage at 2 AM on a Tuesday.
What to do right now during an outage
If the issue might be on your side
- Check your internet connection by loading any website on the same network
- Restart the payment terminal or reload the checkout integration
- Try a different device or switch to a mobile hotspot to rule out local network problems
- Confirm whether the issue affects all payment methods or just one provider
If the issue is at your processor
Check your processor’s public status page first. Most major processors publish real-time status. Downdetector also tracks outage reports across processors and card networks in real time.
Activate backup payment methods if you have them. Digital wallets (Apple Pay, Google Pay) run on separate authorization networks and may be unaffected by a processor-level outage. For in-person transactions, cash is an option. For card-not-present environments, communicate honestly with customers about the delay and offer to complete the transaction once service restores.
Document every affected transaction for reconciliation.
Warning: Check your merchant agreement before using manual card imprinters. Some agreements require authorization for every card transaction, and unauthorized manual transactions may carry full fraud liability under those terms.
How outages impact revenue beyond the failed transaction
The revenue loss from a payment outage accelerates faster than most businesses expect.
Key stat: Payment outages put $44.4 billion in U.S. retail and hospitality sales at risk annually. Between minutes 8 and 13 of an outage, businesses collectively lose $1.2 billion per minute. By minute 23, cumulative losses reach approximately $5.3 billion, roughly 70% of all at-risk revenue (Dynatrace/FreedomPay/Retail Economics, January 2026).
Consumers abandon a purchase after waiting about seven minutes, according to a 2026 joint study by Dynatrace, FreedomPay, and Retail Economics (Payments Dive, January 2026). The average outage lasts two hours. A typical processor outage outlasts consumer patience by a factor of 17.
The immediate transaction loss is only part of the picture. According to PwC research, 32% of customers will leave a brand they love after a single bad experience (PwC, “Experience Is Everything”). After the 2023 Square multi-hour outage, Wolfe Research analyst Darrin Peller estimated 1 to 2% additional customer churn in the following quarter.
For SaaS and platform businesses, the cost compounds. If your platform processes payments on behalf of your customers, an outage at your processor level affects every one of your customers’ end users simultaneously. The reputational exposure isn’t limited to your own buyers; it extends to every business that depends on your infrastructure.
Why relying on a single payment provider creates business risk
Most businesses understand this is a risk. Most haven’t addressed it.
A 2025 survey of 50 enterprise e-commerce leaders by BR-DGE (vendor-sponsored; note the small sample size) found that 92% had experienced payment outages or disruption in the previous two years, and 50% reported losses between £1.1 million and £10 million from those failures (FF News, November 2025). Despite those losses, 71% of the merchants surveyed ran the majority of their transaction volume through a single primary provider.
Only 32% had automated backup routes. The other 68% rely on manual intervention to reroute transactions when a processor goes down, which means someone has to notice the failure, reach the right contact, activate a backup, and test it while customers are already leaving.
15% of U.S. retail and hospitality businesses have no backup payment system at all (Dynatrace/FreedomPay/Retail Economics, January 2026).
Manual failover sounds workable until you time it against the revenue loss curve. Seven minutes is less than it takes to notice the problem, find the right support contact, and activate an alternative. By the time a manual response is complete, most of the revenue exposure has already materialized.
Building payment resilience without engineering overhead
The core requirement for payment resilience is having more than one payment provider and a mechanism to route between them automatically when one goes down. How you get there depends on whether you build it or buy it:
| Factor | Build in-house | Failover platform |
|---|---|---|
| Initial build | 6–12 months (integration, health monitoring, routing logic) | Configuration change, not a development project |
| Adding a backup provider | 3–6 weeks of development per provider | Configuration only |
| Ongoing maintenance | Your team handles API changes, credential management, reconciliation | Platform handles provider API changes |
| Failover speed | Manual intervention (minutes to hours) | Automatic, typically within milliseconds |
Payment gateway failover platforms maintain connections to multiple processors and monitor health continuously. When a provider’s error rate crosses a threshold, transactions reroute to the next available processor, typically within milliseconds, with no action required from your team.
See how the routing and health-check logic works before talking to sales: Payment Gateway Failover covers setup, PSP monitoring, and failover thresholds in detail.
Intelligent payment routing goes further: it evaluates each transaction and selects the provider most likely to authorize it, accounting for processor health, authorization rates by card type, and cost. Multi-processor setups recover 15 to 22% of soft-declined transactions that would otherwise be lost, according to 2024 payment benchmarks from IXOPAY (cited by FlyCode; vendor-sourced figures, treat as directional). For more on how routing decisions enable failover, see a guide to payment routing.
What payment resilience looks like in practice
A business with automated failover doesn’t experience an outage the same way a business without it does.
When the primary processor goes down, the orchestration layer detects the spike in failed authorizations, reaches the configured error-rate threshold, and begins routing transactions to the backup processor. From the customer’s perspective, checkout works. From the operations team’s perspective, an alert fires, and the underlying issue can be investigated without a revenue emergency in progress.
The routing logic sits between your checkout and the payment providers. Customers don’t see it; the checkout flow doesn’t change. For SaaS and platform businesses where your infrastructure serves your customers’ end users, this architecture shift changes the risk profile for every account you support.
Building this from scratch requires integrating with multiple processors, writing failover logic, managing credentials, handling reconciliation across providers, and maintaining all of it as processor APIs evolve. Most engineering teams that attempt this discover the scope grows considerably once they get into it. For an honest look at the build-versus-buy decision, building a future-proof payment stack covers the comparison in concrete terms.
Orchestra connects to 130+ payment providers through a single JavaScript integration. Adding a backup provider is a configuration change, not a development project. For the authorization rate case alongside the resilience gains, see how a multi-processor strategy improves authorization rates.
You cannot prevent a processor outage. You can decide whether your customers see it. Orchestra routes transactions to an alternate processor automatically when a provider goes down, behind the same single integration.
Frequently asked questions
What causes credit card processing outages?
Four main causes: internet connectivity failures, power outages, hardware malfunctions (POS terminals, card readers), and payment processor system failures. Real-world examples include the 2021 TSYS outage, which affected hundreds of thousands of businesses for approximately three hours, and the May 2025 Fiserv outage, which took more than 12 hours to resolve and disrupted Zelle and ACH across major banks.
How long do payment processing outages usually last?
The average outage lasts two hours, according to a 2026 study by Dynatrace, FreedomPay, and Retail Economics. Major processor outages have ranged from about three hours (TSYS 2021) to more than 12 hours (Fiserv 2025).
How much revenue do businesses lose from payment processing outages?
Payment outages put $44.4 billion in U.S. retail and hospitality sales at risk annually. The revenue loss accelerates quickly: between minutes 8 and 13, businesses collectively lose $1.2 billion per minute. By minute 23, cumulative losses reach roughly 70% of all at-risk revenue (Dynatrace/FreedomPay/Retail Economics, January 2026).
What should I do right now if my credit card processing is down?
First, determine whether the issue is local or at your processor. If local, restart equipment and check your network. If it’s your processor, check their status page, activate backup payment options (digital wallets, cash) where possible, and communicate clearly with customers. Document all affected transactions.
How do I tell if there is a payment processor outage?
Check your processor’s official status page first. Downdetector tracks real-time outage reports across major processors and card networks. Most processors also send email or webhook alerts when service degrades.
Can I still accept payments during an outage?
It depends on the cause. If it’s a local network issue, switching to a mobile hotspot may restore service. For processor outages, options include accepting digital wallets on a separate network, taking cash, or using manual card imprinters. Check your merchant agreement before using manual imprinters; unauthorized transactions may carry 100% fraud liability under some agreements.
How do payment orchestration platforms handle gateway failures?
They monitor processor health continuously and automatically reroute transactions to a backup provider when error rates cross a configured threshold, typically within milliseconds. This is the difference between automated failover and the manual processes most merchants currently rely on.
What percentage of customers leave after a payment failure?
32% of customers will leave a brand they love after a single bad experience, according to PwC. A 2023 Square outage was linked to an estimated 1 to 2% additional customer churn in the following quarter, per Wolfe Research analyst Darrin Peller.
