Failover routing
Failover routing is critical for maintaining payment processing uptime. Cardflo automatically detects and responds to acquirer or gateway outages by rerouting transactions to alternative, healthy channels.
This proactive approach minimises downtime and prevents lost sales, ensuring that your payment operations remain resilient under adverse conditions.
- Category
- Routing
- Capabilities
- 10
- Available on
- All plans
The overview
Failover routing serves as a contingency mechanism within the payment orchestration layer to preserve authorisation continuity during technical interruptions.
When a primary acquirer, gateway, or processor experiences a service degradation or complete outage, the routing engine identifies the failure through specific response codes or timeout thresholds. The system then redirects the volume to a pre-configured secondary or tertiary endpoint.
This technical architecture addresses the inherent fragility of single-point processing models by diversifying risk across multiple Merchant Identification Numbers (MIDs) and payment service providers. By decentralising the flow of transactions, merchants can maintain operational stability even when external infrastructure components fail.
The process is typically transparent to the cardholder, as the redirection occurs within the backend authorisation flow before a final response is rendered at the checkout.
Effective failover management requires precise configuration of health checks and priority tiers to ensure that redirected traffic aligns with the merchant's commercial agreements and regional compliance mandates.
How it works
Health monitoring and detection
The system continuously monitors the responsiveness of active acquirers and gateways. It identifies anomalies such as elevated error rates, persistent timeouts, or specific HTTP status codes indicating server-side issues.
If a provider fails to meet predefined performance benchmarks, the routing engine marks that channel as temporarily unavailable to prevent further transaction failures.
Triggering the failover logic
Once a disruption is confirmed, the orchestration layer intercepts incoming authorisation requests intended for the affected provider.
The logic evaluates the transaction metadata, including the BIN, currency, and MCC, to determine which alternative processing path is best suited to handle the redirected volume while maintaining high authorisation prospects.
Dynamic path redirection
Transactions are instantly rerouted to a secondary or tertiary acquirer as defined in the merchant routing table.
This transition happens in real time, ensuring that the payment process continues without requiring the consumer to re-enter their card details or restart the checkout session, thereby protecting the conversion rate.
Automated recovery and rebalancing
The system continues to probe the primary endpoint with low-volume tests or heartbeat requests. Once the service provider demonstrates stability and returns to normal operation, the failover rule is deactivated.
Traffic is then gradually rebalanced back to the primary channel in accordance with the merchant's cost and volume preferences.
Why it matters
Mitigating systemic downtime risks
Even Tier 1 acquirers experience scheduled maintenance or unplanned technical incidents that can halt processing for hours. Failover routing acts as a structural safeguard, ensuring that a merchant's ability to accept payments is not tethered to the uptime of a single partner.
By distributing traffic across a multi-acquirer setup, businesses can insulate their revenue from localised infrastructure failures and maintain a consistent processing profile.
Protecting the customer experience
Technical declines caused by processor instability often lead to cart abandonment, as many cardholders will not attempt a second transaction after a system error. Failover mechanisms resolve these issues in the background, allowing the transaction to clear through a healthy path.
This minimises the occurrence of soft declines related to system connectivity, preserving brand reputation and reducing the customer support burden associated with payment failures.
Use cases
High volume flash sales
During peak traffic events, a primary gateway may become throttled or overwhelmed. Failover routing automatically shifts excess load to a secondary provider, preventing a total collapse of the checkout process.
Cross border processing
When a regional acquirer faces connectivity issues with international schemes, failover rules can redirect the traffic to a global acquirer to ensure that foreign transactions continue to be authorised without interruption.
Scheduled provider maintenance
Merchants can pre-emptively activate failover paths during a known maintenance window of their primary PSP, ensuring that midnight batches and recurring billing cycles proceed through an alternative active channel.
SCA and 3DS outages
If an issuer or ACS provider has latency issues, the system can attempt an alternative route that may have different connectivity to the relevant authentication rails, minimising potential friction.
By the numbers
Typical availability achieved by merchants using multi-acquirer redundancy to bypass individual provider outages, based on industry-standard infrastructure reliability.
The estimated volume of transactions usually lost to technical declines and service timeouts that can be recovered through automated path redirection.
The standard duration required for a modern orchestration engine to identify a gateway timeout and initiate an alternative routing path.
Related terms
Talk to our team about a live rollout on your acquiring stack.
What you get with Failover routing
- Real-time identification of acquirer latency and service degradation through automated heartbeat monitoring systems.
- Categorisation of decline codes to distinguish between terminal failures and temporary processing interruptions.
- Configurable prioritisation of secondary and tertiary endpoints based on settlement speed and scheme fees.
- Seamless redirection of Merchant Initiated Transactions to ensure recurring billing cycles remain uninterrupted.
- Automated rebalancing of traffic once the primary processing path returns to a healthy state.
- Reduction of soft declines caused by technical timeouts or gateway-level communication errors.
- Support for multi-MID setups to diversify risk across different geographical banking entities.
- Customisable timeout thresholds to prevent excessive wait times during the authorisation attempt.
- Granular logging of failover events for post-incident analysis and service level agreement monitoring.
- Integration with smart routing logic to maintain cost-efficiency during emergency redirection periods.
A short scoping call, then a written plan for your MIDs.
Questions about Failover routing
What is the difference between failover routing and load balancing in payments?
Load balancing distributes transaction volume across multiple providers simultaneously to optimise performance or costs. Failover routing is a reactive mechanism that only redirects traffic when a primary provider fails.
While load balancing is a standard operational state, failover is a recovery event triggered by specific error thresholds or downtime. In a sophisticated payments stack, both are used together to ensure both efficiency and resilience.
Load balancing manages the daily flow, while failover provides the safety net.
How does the system distinguish between a valid card decline and a processor outage?
The failover engine analyses response codes returned by the gateway or acquirer. Valid declines, such as insufficient funds or an incorrect CVV, do not trigger a failover because the issue lies with the cardholder or issuer.
However, codes indicating a system error, timeout, or a refusal to connect are flagged as infrastructure failures. The system only initiates a reroute when the error is identified as being external to the cardholder's account status.
Can failover routing impact my interchange rates or scheme fees?
Redirecting traffic to a secondary acquirer may result in different commercial terms. If the secondary provider has a higher interchange-plus-plus margin or different scheme fee structures, the cost per transaction might increase during the failover period.
However, this marginal cost is generally considered preferable to the total loss of a transaction. Merchants should configure their failover paths to prioritise the next most cost-effective provider in their hierarchy.
Does failover routing require the customer to re-enter their payment details?
No, if the merchant uses a central vault or payment orchestration layer, the card data is captured once and held securely.
If the first authorisation attempt fails due to a technical error, the system use the existing token to attempt the transaction with the secondary provider.
To the cardholder, this may appear as a slightly longer processing time, but the interface remains consistent and no secondary data entry is required.
What happens to 3D Secure authentication during a failover event?
If 3DS authentication has already been successfully completed, the authentication payload must be compatible with the secondary acquirer for the failover to succeed. If the interruption occurs before 3DS is finalised, the system may need to restart the authentication flow with the new provider.
Advanced orchestration platforms manage these handovers to ensure that the SCA requirements of PSD2 are still met during the redirection.
How quickly can the system detect a failure and switch to a new acquirer?
Detection usually happens within milliseconds or after a defined number of consecutive failures. Merchants can set thresholds, such as three consecutive 5xx errors or a timeout exceeding five seconds.
Once the threshold is met, the switch is instantaneous for all subsequent transactions. This rapid response is critical for maintaining high-volume checkout flows where even a few minutes of downtime can result in significant revenue loss.
Ready for velocity?
Tell us about your business. We'll match you with the right acquiring partners and the right route, typically inside a week.