How Much Does POS Downtime Cost a Restaurant?

Every restaurant owner has a moment they can point to. The card reader freezes mid-transaction on a Friday night. The internet drops during a patio rush. The kitchen display goes dark and someone has to run tickets by hand. For a few minutes, or a few hours, the system that runs the whole operation simply stops.

What happens next tells you a lot about how a restaurant is built. Some operations keep serving. Others grind to a stop, and that’s when the real cost of downtime starts to show.

What actually happens when a POS goes down

The first cost you notice is the lost sale. A guest can’t pay, the queue grows and people start leaving. That’s the most visible part of an outage, but not the most expensive.

Meanwhile, staff can’t do their jobs properly. Servers switch to handwritten orders, the kitchen loses its ticket queue and managers end up fixing the router instead of managing the floor. You’re still paying for every hour, without the usual revenue.

Waiting guests start walking away. They hold the restaurant responsible, and first-time visitors rarely give you much leeway. If the outage lands on a big night or holiday weekend, it affects the people you need to impress most.

In the kitchen, the rhythm falls apart. A mid-rush KDS failure disrupts the sequencing that keeps orders moving without delays or over-firing. Teams on tight rosters feel the impact straight away.

The Rogers outage: a real test of what downtime costs

Canada already had a national stress test for this question. On July 8, 2022, a maintenance update at Rogers Communications took down mobile and internet service for more than fifteen hours across much of the country. Interac, which many Canadian banks route through, went offline along with it. Debit stopped working at the till. E-transfers stopped moving. ATMs went dark.

Cash-only signs appeared outside restaurants and cafés across Ontario within hours. In Toronto, a café owner estimated she lost a few hundred dollars in one afternoon when guests who couldn’t pay went elsewhere for Wi-Fi. An independent coffee shop in Saskatchewan scrambled to find payment workarounds mid-shift. The disruption crossed regions and restaurant types, hitting any business whose payments relied on the network.

Dan Kelly, president and chief executive of the Canadian Federation of Independent Business, later said small businesses lost thousands of dollars that day. Many couldn’t process online orders, delivery apps or card payments. The outage is a useful reference point because the POS wasn’t the problem. It showed that a restaurant’s exposure includes every system it relies on to take a guest’s money.

It isn't always the internet provider's fault

Rogers showed what happens when the network underneath every payment system fails at once. But some of the most disruptive outages in recent years came from the POS and payment vendors themselves, which is a harder problem for an operator to plan around because it isn't solved by switching internet providers.

In September 2023, configuration changes overloaded Square’s internal DNS servers, disrupting its services for about 15 hours. A Cincinnati pizzeria counted at least $651 in lost sales and almost $90 in lost tips before giving up. That didn’t include regulars who tried to order online, couldn’t and never called. A San Antonio café closed early because it had no way to take payment. Square confirmed that a faulty software rollout, not a breach, caused the outage and said it would expand offline payments as a result.

Similar failures keep happening. Earlier in 2023, a ransomware incident disrupted some NCR Aloha cloud services and administrative functions. NCR said in-restaurant purchases and transactions continued, but affected customers had reduced access to management tools. The previous month, hackers disrupted a major foodservice distributor and forced it to pause deliveries. A configuration change took McDonald’s POS systems offline in several countries in March 2024, shutting some locations through busy breakfast and lunch periods. Years earlier, a failed overnight update caused the same problem at Starbucks stores across the US and Canada. With no way to take payment, some gave the coffee away.

Cloud infrastructure isn’t immune either. When a major cloud provider experiences an outage, multiple restaurant technology vendors can be affected at the same time.

During these incidents, restaurants that were fully dependent on a live connection to a vendor's cloud, no matter how reliable that vendor usually is, stopped. Offline workflows and local fallbacks can keep core service moving when online ordering, reporting or integrations are unavailable.

What resilient systems look like

The difference between a restaurant that rides out an outage and one that loses the night often comes down to a few decisions made long before anything breaks.

An on-premises system with a local server can often keep taking orders without internet, though updates may be slower and the hardware can still fail. Cloud-based systems are easier to maintain, but their outage performance depends on what remains available offline. A hybrid setup can combine cloud reporting and multi-location visibility with local tools that keep core service moving.

That makes a real difference during service. If the KDS loses its connection but the kitchen printer keeps firing tickets, the back of house keeps moving. If payment terminals accept cards offline and settle them later, staff don’t have to tell an already seated table it’s cash only. This doesn’t eliminate downtime. It changes what downtime costs.

It also raises a question operators often leave too late: how is the platform built? A system designed with offline mode built into its hardware and payments rather than added later, handles pressure differently from one that wasn’t designed around reliability. You won’t always spot that in a sales demo. You will when the internet drops on a Saturday night.

Building downtime into the plan

Most restaurants only calculate the cost of downtime after a bad shift. It's more useful to estimate it before the next outage happens.

Start with your average peak-hour sales and estimate how much revenue you'd lose during a typical disruption. Then add the direct costs of getting service back on track, including refunds, comps, overtime, remade food and recovery work.

The cost can carry into the next shift and beyond. Some guests won’t return after a poor experience, negative reviews can deter future customers and a disruption can hurt the rest of the day’s sales. Those effects are hard to price, but they still add to the cost of an outage.

Staff training should be part of the resilience plan from the start. When the team knows the manual backup process, who can comp a delayed order and how to take written orders during a rush, they can recover in minutes instead of spending most of the shift figuring it out.

Reliability should also carry more weight when choosing a system. Monthly prices and features are easy to compare, but resilience only becomes obvious when something fails. Ask each vendor about its uptime record, exactly how offline mode works and what staff can still do without a connection. The strongest answer may come from the platform that keeps trading while a competitor nearby is stuck taking cash.

The strategic takeaway

Downtime isn't a technology problem that happens to restaurants. It's an operational risk that restaurants can plan for the same way they plan for a bad weather weekend or a staff call-out on a Saturday. 

The Canadian restaurants that come through an outage with the least damage tend to share two things: a system built with some form of offline resilience, and a team that's rehearsed what to do when the screen goes dark. Neither guarantees the internet stays up. Both change what happens when it doesn't.

If you're evaluating how your current setup would hold up, Toast's guide on what happens to restaurant operations when the internet goes down walks through the practical side in more detail.