Retail Failover Case Study: Keeping Payments Live

A card terminal that cannot reach its payment platform does not look like a connectivity problem to the customer. It looks like a shop that cannot take payment. In this retail failover case study, we examine a representative multi-site retailer that replaced a single fixed-line dependency with mobile backup designed to keep essential services operating when the primary connection failed.

The issue was not constant broadband failure. It was the impact of occasional failure at exactly the wrong moment: a damaged local cable, an ISP fault, maintenance work, or a router problem during a busy trading period. A few minutes offline could mean abandoned baskets, manual workarounds, frustrated staff and a queue at the till.

For retailers, failover is not about adding another internet connection for its own sake. It is about deciding which systems must remain available, how quickly they need to recover, and which network path can be trusted when the usual one disappears.

The retail problem: one outage, several operational failures

The retailer operated small and medium-sized sites with broadband serving payment terminals, back-office systems, staff Wi-Fi, stock devices and guest access. This arrangement was convenient, but it treated very different services as though they carried the same priority.

When the fixed connection dropped, payment traffic competed with every other connected device - or stopped altogether. Store teams could sometimes use mobile phones as temporary hotspots, but that was not a repeatable operating model. It relied on an available handset, adequate signal from one carrier, a member of staff who knew what to do, and enough time to reconnect devices while customers waited.

The business set three clear requirements. Payment processing had to remain available wherever possible. Store managers needed a simple setup with no complex intervention during an incident. Central operations needed visibility of data use and connection status without calling each branch for an update.

A single-network SIM was considered, but it created a familiar weakness: good coverage from one carrier at a site does not guarantee good coverage during congestion, local disruption or a change in indoor conditions. The retailer needed a failover route that could make a better network decision in real time.

Retail failover case study: the design

The solution used a business-grade cellular router at each location, connected to the store network as a secondary WAN path. The router contained a non-steered multi-network data SIM, allowing it to connect automatically to the strongest available major carrier rather than remaining tied to a preferred network.

Under normal conditions, the fixed broadband line handled day-to-day traffic. The cellular connection remained ready in the background. When the primary line failed health checks, the router moved designated traffic onto mobile data. When fixed service recovered and proved stable, traffic returned to the normal path.

That last detail matters. Failover should not simply detect whether a cable is plugged in. A broadband circuit can appear connected while DNS resolution, payment platform access or general routing has failed. Meaningful health checks test the services the retailer actually depends on.

The network was also segmented. Payment terminals and essential operational devices were placed in a priority group, while guest Wi-Fi and non-essential updates were restricted during a failover event. This protected the available mobile-data capacity for transactions, stock look-ups and communications that kept the shop functioning.

Why multi-network connectivity changed the outcome

In a conventional backup setup, a cellular router may be technically online but still struggle because its one assigned network has weak indoor coverage or limited capacity at that moment. That is an uncomfortable form of false reassurance.

Multi-network connectivity gives the router more options. It can attach to the strongest suitable available network at the site, helping to reduce the chance that a fixed-line failure becomes a complete connectivity failure. It is not a promise that mobile service will be perfect in every location - building materials, local signal conditions and carrier availability still matter - but it removes a major single-carrier dependency.

For a retailer with branches in different towns, retail parks and high streets, that flexibility is commercially useful. A network that performs well at one shop may be the weaker choice at another. One SIM model can support the estate without forcing operations teams to source and manage separate carrier arrangements site by site.

What happened when the primary line failed

During a weekday broadband incident, the router detected repeated failed connectivity checks and triggered failover. Essential traffic moved to the cellular connection automatically. Store staff did not have to swap SIMs, enable a phone hotspot or contact technical support before taking the next payment.

Payment terminals continued to process transactions, while guest Wi-Fi was paused and background device updates were held back. The branch remained open and trading. From the customer’s perspective, there was no change at the checkout.

Central teams could see that the location was operating over mobile data, review consumption and confirm when the primary connection had returned. This changed the incident response. Rather than asking whether the shop was down, operations could focus on the root cause and make a measured decision about escalation.

The practical value was not limited to one event. The retailer now had a tested process for an outage that could happen anywhere in its estate. That repeatability is what turns a backup connection into operational resilience.

The decisions that made failover work

The SIM was only one part of the result. Retail failover works best when the technical design reflects how the store operates.

First, the retailer defined critical services in advance. Payment terminals came first, followed by devices needed to fulfil orders and manage stock. A shop may also need CCTV connectivity, digital signage controls or staff communication tools, but not every service should receive equal access to backup capacity.

Second, the team tested real failover behaviour. They disconnected the primary WAN, measured recovery time, processed test transactions and checked which devices moved across. They also tested the return to fixed broadband. Failback can be disruptive if it happens too quickly during an unstable line, so a sensible stability threshold was configured before traffic moved back.

Third, they assessed mobile signal where the equipment would actually sit. A coverage map is useful for planning, but it cannot fully account for a comms cupboard, metal shelving, reinforced walls or a basement till area. In a challenging site, repositioning the router or fitting suitable external antennas may make a material difference.

Finally, they managed data with intent. Retail payment traffic is typically modest, but software updates, cloud backups, guest usage and media-heavy devices can consume far more than expected. Restricting non-essential traffic during failover keeps costs predictable and protects the service that matters most.

Where retail failover needs a different approach

Not every store needs the same setup. A single independent retailer may use a compact router and keep its configuration focused on payment terminals. A multi-site operator may require central policies, remote monitoring, traffic rules and standardised deployment across hundreds of locations.

The trade-off is between simplicity and control. A basic backup setup is quick to deploy, but it may offer limited visibility and weaker traffic prioritisation. A managed design takes more planning, yet it gives IT and operations teams clearer control when a site moves onto mobile data.

There are also cases where cellular should be the primary connection, not merely the fallback. Pop-up shops, seasonal concessions, outdoor retail, temporary units and sites awaiting a fixed installation can operate effectively on mobile data from day one. In these deployments, multi-network access becomes the main route to continuity rather than an emergency option.

Wave Connect supports this model with non-steered multi-network data SIMs and a central management platform, giving retail teams a practical way to deploy, monitor and manage connectivity across locations.

A better question than “Do we have backup?”

The useful question is not whether a shop has a backup connection. It is whether that backup can keep the right services live, switch automatically, find viable coverage at the specific site and be managed without a last-minute scramble.

Test the answer before the next outage. Put the router where it will be installed, simulate a fixed-line failure during a quiet period, run a transaction and check what the branch team can see. The best failover plan is the one customers never notice - because the checkout keeps moving.