Skip to content

Field notes · POS

When the internet goes down: what your POS actually does next

Editor's field notes, drawn from three years of conversations with independent restaurant operators. As the verified community grows, these notes will incorporate direct member signal — with consent, always anonymized.

By The Editors, OrderBridgePublished

Restaurant worker holding a handheld POS device at the service counter.
Illustrative editorial image · OrderBridge field library

Recurring signal

The pattern

A cloud POS needs the internet to sync, report, and take some payment types — but a well-configured one keeps selling through a short outage. The recurring operator surprise was not that offline mode exists. It was learning who is financially responsible for a card payment that gets declined after the connection returns.

Every major platform's own documentation says the same thing in slightly different words: the restaurant, not the platform, carries the risk on a stored offline payment that is later declined, expired, or disputed. Operators who planned around that fact lost far less on their worst outage night than operators who discovered it during one.

Pull-quotes are editorial reconstructions of recurring operator accounts. Identifying details are changed.

Offline mode buys time, not certainty.

Toast's published support documentation describes devices entering Offline Mode automatically after about 40 seconds without connectivity, after which a visible banner explains what functions remain available. With background card processing enabled, a Toast restaurant can keep taking card payments with tips, printing, and opening cash drawers — card data is encrypted and stored locally until the connection returns, at which point it is submitted for authorization. Square's current documentation describes a comparable capability, with offline payments enabled by default and a per-device cap on stored transaction value.

What stops working during an outage varies by platform but recurred consistently in operator reports: online-ordering intake, gift cards, loyalty redemption, menu edits, and end-of-day shift closing typically pause until the connection is restored, even while core order-taking continues.

"The banner said we could keep selling. It didn't say our online orders were queued up on a server somewhere, about to arrive all at once the second we reconnected."

— general manager, delivery-forward restaurant (details changed)

The restaurant owns the risk on a stored card payment.

Toast's platform documentation states plainly that the restaurant is responsible for any declined, expired, or disputed payment accepted while in offline mode — and warns that some card-network authorizations can expire in as little as 24 hours, recommending devices return online within one to three days at the outside. Square's documentation carries an equivalent line for its own offline-payment window, which runs 24 to 72 hours depending on the device, with a hard expiry after which stored payments cannot be retrieved or reprocessed at all.

This liability structure was consistent across every major platform reviewed: none of them absorb the risk of a card that ultimately does not clear. Operators who understood this treated a prolonged outage as a live financial exposure to manage actively, not a technical inconvenience to wait out passively.

"We assumed the POS company ate the cost if a card didn't clear later. Reading our own platform's documentation made clear that assumption was simply wrong."

— owner-operator, full-service restaurant (details changed)

A cellular failover turned most outages into non-events.

The single most repeated recommendation across operator accounts and published guidance was the same: a cellular (4G/5G) failover router, automatically taking over when the primary internet connection drops, at a reported cost of roughly $30 to $50 a month plus a one-time hardware cost in the $400 to $800 range. That investment was consistently described as eliminating the large majority of outage impact — turning a scenario that once meant a stressful manual fallback into one the staff barely noticed.

Operators without a failover in place, comparing that modest monthly cost against the revenue at risk during even one lost Saturday night of card volume, generally treated the decision as settled once they ran the comparison explicitly rather than deferring it as a future project.

"The cellular backup cost less per month than one bad Saturday night without it. We just hadn't done that specific comparison before the first real outage forced us to."

— operator, quick-service restaurant (details changed)

A cash float remained the simplest backstop of all.

Even restaurants that had gone nearly cashless reported keeping an emergency float — commonly $500 to $1,000 — specifically for outage nights, on the reasoning that some guests would rather pay cash on the spot than have a card payment's fate depend on a reconnection that might not happen for hours. The float only worked, operators noted, when the restaurant could actually make change for it — a detail easy to overlook in a mostly-cashless operation until the moment it mattered.

Staff training for outage scenarios recurred as a separate, low-cost habit that paid off disproportionately: a short, rehearsed script for explaining a card-authorization delay to a guest, and a clear internal decision tree for when to switch to cash-only, prevented the confusion that made a technical outage feel like a service failure to the guest in front of the register.

"The outage itself lasted twenty minutes. The confusion at the register, because nobody knew what to say to guests, lasted the whole shift — until we actually wrote down what to say."

— shift manager, full-service restaurant (details changed)

Failover cost versus one lost Saturday

The math

A cellular failover router runs an illustrative $400 to $800 one-time hardware cost plus $30 to $50 a month. A single Saturday night's card volume for a mid-sized restaurant can easily exceed $3,000 to $5,000 — the amount at real risk if a multi-hour outage forces a scramble to cash-only mid-service, on top of any offline payments that ultimately fail to clear.

Cellular failover cost versus one Saturday night's card volume at riskA cellular failover router costs roughly four hundred to eight hundred dollars one-time plus thirty to fifty dollars monthly. One Saturday night's card volume at a mid-sized restaurant can exceed three thousand to five thousand dollars, the amount at risk during a multi-hour outage.A monthly subscription versus one bad nightIllustrative comparison · mid-sized full-service restaurantCELLULAR FAILOVER$400–800 + $30–50/mo.ONE SATURDAY AT RISK$3,000–5,000+The backup usually costs less than the exposure it removes.
Illustrative figures compiled from published cellular-failover pricing and typical restaurant card-volume ranges. Actual exposure varies by restaurant size and outage duration.

Context from operator conversations

The tools that came up

Featured in operator conversations

Toast

Toast's offline-mode documentation is frequently the reference point operators cite when explaining outage risk to staff, since its background-processing settings and authorization-expiry warnings are published in detail.

Learn more →

Affiliate partner. Full disclosure at /disclosure.

Source notes

Public documents behind the field notes

  1. Toast support: preparing to operate in Offline Mode.
  2. Toast platform guide: offline card payments.
  3. Square current offline-payments documentation on device-based transaction limits and expiry windows, reviewed 2026.

Source review: July 17, 2026. Offline-mode capabilities, limits, and expiry windows are set by each platform and change over time.

Verified operators only

Add your field notes.

Membership is free and always will be. Applications are reviewed weekly. Vendors are declined.

Request to join →

From the Editors

Sunday field notes.

One useful note from the room each week. No feed noise, no vendor spin, and an unsubscribe link in every edition.