Site icon AfricaBusiness.com

Retail POS Downtime: A Practical Guide to Payment Fallback

Keeping card payments moving during an outage requires working equipment, an available authorisation route and a clear plan for matching payments to sales afterwards.

By Rory Bosman, Chief Sales & Marketing Officer, and Gary Bowers, Product Owner, Ecentric Payment Systems

Retail point-of-sale downtime creates two operational tasks: keeping eligible payments moving and maintaining a reliable record of what was sold. A continuity plan needs to address both.

A functioning card terminal may allow a retailer to accept payment while its till software is unavailable. But that payment record does not automatically recreate the itemised sale, update inventory or complete reconciliation.

For South African retailers reviewing their arrangements, the starting point is to identify which part of the operation has failed. “The POS is down” can describe several problems, each requiring a different response.

Match the fallback to the failure

A till application, a payment terminal, the store’s connectivity and the processing infrastructure perform different functions. Their dependencies determine whether trading can continue.

Ecentric’s written clarifications describe the following arrangements:

Failure Possible continuity route Important limitation
POS software or a required server becomes unavailable Staff may use Ecentric’s standalone Fallback Payment app on a functioning terminal. The terminal still needs an available online payment route. Pricing, item records and other till functions require separate arrangements.
The primary network connection is lost A suitably configured terminal may connect through its backup mobile SIM. Mobile coverage and the payment infrastructure must remain available. Other store functions depend on which network links and servers still work.
The acquiring service becomes unavailable Merchants configured for backup acquiring may have transactions redirected to another acquirer through Ecentric. This requires prior configuration. It should not be assumed to cover every failure elsewhere in the processing chain.
The physical terminal fails Staff need another functioning, supported terminal. An application installed on the failed device cannot replace working hardware.

An alternative connection and an alternative acquirer address different dependencies. Retailers should establish what their provider’s continuity arrangement covers, including failures affecting the payment switch or other infrastructure shared by the normal and fallback routes.

A centrally managed change of acquirer also does not remove the terminal’s need to communicate with the payment service.

Standalone payments still need online authorisation

In the configuration described by Ecentric, fallback means using a standalone payment application with an available connection. It does not mean authorising cards offline or storing transactions for submission later.

Each payment is submitted for online authorisation through the Ecentric switch and an acquirer. Declines are returned at the terminal, and the usual card-scheme and acquiring rules continue to apply.

That distinction matters when assessing a store’s ability to trade through a disruption. If neither the primary connection nor the alternative connection can reach the required payment services, this form of fallback cannot provide online authorisation.

Retailers should confirm which payment methods and transaction types their configuration supports, how staff activate the fallback process and what happens when the alternative route is unavailable.

Preserve the payment reference before retrying

An uncertain payment outcome requires particular care. A missing response does not, by itself, tell the cashier whether the payment was approved or declined.

The POS software generates the transaction reference before sending the payment request. If the response is missing, the POS can use that same reference to look up the original transaction through Ecentric’s API, provided the required systems are reachable.

This supports duplicate-payment prevention when the same transaction is correctly identified. Staff still need a documented procedure for checking an uncertain outcome before initiating another payment.

During an outage, the relevant checking route may itself depend on which systems remain reachable. The procedure should therefore explain who investigates, where the payment reference can be found and what staff should do while confirmation is pending.

A payment record is only part of reconciliation

Ecentric says fallback transactions are identified on its platform and appear in its portals and settlement files. Those records support the retailer’s recovery process.

Matching them to itemised sales and stock movements is a separate integration task.

For POS systems using a cloud-based central system, Ecentric says notifications can be provided through webhooks or a secondary payment advice. Retailers should confirm what those notifications contain and how their own systems use them.

For POS systems relying on an on-site server, Ecentric says automatic matching after recovery remains an area under development. Complete automatic restoration of sales and inventory records should therefore not be assumed.

The practical question is what information survives in each system and how the retailer brings those records together. An approved card payment, a settlement entry and an itemised sale need to be matched without creating another charge.

A hypothetical South African retail example

Consider a South African grocery store whose till software loses access to a required server. Its payment terminal remains powered and can reach the payment service over mobile connectivity.

Before continuing, staff need an approved way to establish prices and record the goods being sold. They then use the standalone payment application and retain the payment reference alongside the sale record.

If a payment outcome is uncertain, staff follow the agreed checking procedure before attempting another charge.

When the till system becomes available, the retailer matches fallback payment records to the recorded goods and checks them against the relevant settlement information. The sale and stock records must then be completed without charging the customer again.

This is an illustrative scenario, not a documented customer deployment or a measured recovery result. Its purpose is to show why accepting payment and restoring retail records require connected procedures.

Assess migration and compliance for the actual deployment

Application-based integration can make supported terminal changes easier to manage. Ecentric says its approach allows a POS integration to be reused across supported devices.

Compatibility still needs to be checked. A replacement device may introduce a different Android version, and the merchant’s POS application may require an update. The supported hardware and applicable approvals also matter.

Android alone does not establish reduced compliance requirements. The PCI Security Standards Council explains that a PCI-listed point-to-point encryption solution can reduce the number of applicable PCI DSS requirements, while PCI DSS remains applicable.

Retailers should obtain the relevant solution listing and confirm how their deployment meets its conditions. A supported or approved terminal on its own does not establish that the complete implementation qualifies.

The cost assessment should include devices, connectivity, integration, support, training and reconciliation work. The material supplied for this article does not establish a quantified saving from migration.

Six questions to ask a payment provider

Before relying on a fallback arrangement, retailers should ask:

  1. Which failures have been tested? Distinguish till software, local networking, internet connectivity, acquiring services, payment-switch infrastructure and terminal hardware.
  2. What must remain available? Confirm power, mobile coverage, supported payment methods and the services required for online authorisation.
  3. How are uncertain outcomes checked? Establish how staff retrieve the original transaction reference and confirm its status before retrying.
  4. How are payments matched to sales? Identify the records available during disruption, the integration used after recovery and any manual reconciliation.
  5. Which configurations and approvals apply? Confirm supported devices, operating-system versions and the basis of any claimed PCI DSS scope reduction.
  6. Who owns recovery? Agree staff procedures, escalation contacts and how recovery time, reconciliation effort and costs will be measured.

Payment continuity is a property of the complete operating arrangement. Hardware, connectivity, authorisation, staff actions and record matching all contribute.

A useful continuity test therefore ends when the retailer can explain what was sold, confirm the payment outcome and reconcile the records—not simply when a terminal accepts a card.

About the authors

Rory Bosman is Chief Sales & Marketing Officer and Gary Bowers is Product Owner at Ecentric Payment Systems, the provider of the fallback solution discussed in this article.

Rory Bosman, Chief Sales & Marketing Officer at Ecentric Payment Systems.
Gary Bowers, Product Owner at Ecentric Payment Systems.

Author portraits supplied by Ecentric Payment Systems’ PR team.

Sources and scope

Product-specific descriptions reflect Ecentric’s written explanations. Capabilities depend on the merchant’s configuration and integrations. The retail example is hypothetical; the article does not report an independent field test.

Exit mobile version