When a POS outage hits during service, your team needs a clear 30-minute playbook, not panic. This guide walks through exactly what to do, who does what, and how to keep guest trust, order accuracy, and closing numbers in balance.

The POS goes quiet at 7:03 p.m. on a Friday, and for about six seconds everyone keeps moving as if nothing changed. Then the line notices the pause. The host still has two open windows, the kitchen is waiting on ticket count, and your cashier is staring at a frozen screen that used to be the source of truth. In that exact 60-second window, this is where confidence is won or lost.

At this stage, teams usually make one of two mistakes: they sprint for buttons and settings, or they freeze and hope the terminal wakes up by itself. Neither helps. Most POS vendors queue offline transactions locally and send them when connectivity returns, and that gives you a useful strategy: protect the data you still can, keep the guest flow moving, then reconcile with discipline when the system returns. A rescue is less about speed tricks and more about role clarity and short memory.

The first thing to know about outages

A POS outage is a systems problem, not a character problem. People do not perform poorly because they are careless. The room just gets harder to coordinate and the normal feedback loops stop. The key is to replace dependence on one broken screen with a tiny second system you can control: a short manual flow, a visible command list, and one person who guards accuracy.

Minute 0 to 10: protect guests and stabilize the line

During these first ten minutes, your objective is clear: keep the restaurant moving and prevent duplicate work. Start with a small statement to staff, not a lecture. Something plain: "Offline. Manual mode for tickets and tender starts now. No one restarts devices yet." People follow simple language better than panic noise.

• Call one coordinator and one runner. The coordinator says what is happening. The runner keeps guests informed and moves takeaway handoffs.
• Stop touching payment terminals unless someone is already mid-transaction. If a card swipe is hanging, let it time out and move on.
• Switch to paper tickets immediately. Each ticket should have guest count, order items, modifications, and a rough timestamp. It is boring work. It is also your backup for later reconciliation.
• Tell guests what happened in one sentence: "Our POS is offline, but your order is still being tracked." No overexplaining, no apology marathon.

If your store has more than one line, split by station rather than by urgency. That keeps one issue from paralyzing the whole front house. A host can still greet and seat, and a runner can confirm specials and order timing while cashiers handle offline ticket numbers. The goal is not elegance. The goal is reduced confusion.

"Most teams lose more revenue in the first five minutes than in the next ten. Not because the issue is fatal, but because everyone assumes someone else is already in control."

Minute 10 to 20: lock down roles, payments, and credit policy

Once the first wave of customers is no longer collapsing into chaos, assign three people to three roles, and keep those roles fixed for the next ten minutes.

Role 1: Order controller

This person owns the manual order pad, tracks seat count, and makes sure every manual ticket gets written once. If a duplicate order appears, they own that correction. If your operation includes dine-in and pickup, this role marks channel and promised time on each ticket.

Role 2: Payment gatekeeper

This person handles non-card payment instructions, checks cash drawer discipline, and confirms split-check requests on a single line in the pad. If card payments are restricted offline, be explicit with guests that card settlement is pending if required by your local process. Do not promise a real-time charge that your stack cannot confirm.

Role 3: Reconnect watcher

This role tracks network restoration attempts and reports the exact minute status changes. If reconnect happens at minute 13, great. If it happens at minute 18, everyone still runs from the same manual plan and the order count remains coherent.

When the line starts to normalize, update a simple board: total tickets, open items, known issues, and any unresolved payment exceptions. A dry-erase list sounds old school, but it gives the whole team one version of truth.

Small but important rule: if a card is declined offline, your team should not keep re-trying blindly. Ask for alternate payment options and mark it clearly. One clean retry with a clear reason is better than five random attempts and a confused guest waiting again.

Minute 20 to 30: protect your books before the system wakes up

At this point, you are no longer just surviving. You are preparing to return to clean accounting records. The first 10 minutes bought speed, the second 10 bought structure. These last ten prevent quiet errors.

1. Count tickets on paper against payments received by cash, card intent, and gift balance. You want total guest count and total gross value to match your manual log, not the POS.
2. Mark any order with substitutions, custom asks, or allergy flags in bold pencil. They are easy to lose during manual writeups and expensive to recover later.
3. Create a "hold" section for unresolved transactions where connectivity failed during authorization. No one should close the night shift without this list resolved.
4. Before reconnect, verify the manual log is complete for the outage window. Missing seat count, missing checks, and missing notes all become disputes later.

When a service app reappears, some teams panic and start reentering everything at once. Slow down. Reconcile station by station. First reopen dine-in tickets by time block, then refill orders that had pending card intent, then reopen held notes. This takes a little longer, but it is far less painful than fixing a full-day mismatch.

Where recovery fails most often

If you have to run this drill repeatedly, your issue is usually not the POS. It is the transition between systems. These are the three most common failure points:

• People changing roles mid-drill. A host who starts as queue guide and then starts entering tickets creates overlap and missing entries.
• Unclear language. "System is down" is not enough. Use specific action verbs: write tickets, hold card exceptions, keep manual log.
• Hidden communication. If only the floor manager knows the plan, one table near pickup can derail the whole shift. Everyone in service needs the one-line rule.

There is a cultural reason these mistakes repeat: outages are rare, and rare tasks get forgotten. That is why this should be a recurring drill, not a heroic rescue memory. If a team can handle a 30-minute outage calmly once a quarter, they can handle it without drama later.

A practical script you can copy tonight

If you want one ready-to-run version, read this with your team during shift huddle.

Outage Script (30 Minutes):

1. Announce: "POS offline, switch to manual ticket flow, keep calm."
2. Assign coordinator, runner, payment gatekeeper.
3. Record every ticket with seat count, time, modifications, and payment status.
4. After 10 minutes, verify duplicate prevention and clear an "exception" list.
5. At 20 minutes, do a fast board check: total tickets, total payment intent, unresolved auths.
6. At 30 minutes, freeze data, compare paper log with any recovered POS data, then settle reopen order by order.

If this feels too intense, scale it. A smaller restaurant can do the same routine with two people and one whiteboard. A larger venue can assign a separate reconciliation lead and still follow these same time blocks.

Why this keeps margins and trust from leaking

The cost of an outage is usually measured at the register, but there are hidden costs behind that number: wrong payouts, guest mistrust, and team burnout. A predictable 30-minute response gives teams confidence. Guests may tolerate a brief outage. They do not tolerate silence, confusion, or repeated corrections.

Small teams recover fastest when expectations are explicit. You cannot over-communicate during a live incident, but you can over-communicate before it happens. Put this drill on your wall, test it monthly, and make sure your newest hires can repeat it in under two minutes.

For teams already using M&M POS, training this 30-minute script into shift routines can reduce both line stress and reconciliation errors. You can download M&M POS for the setup and recovery documentation that supports offline workflows and team training materials.

Direct URL: https://mmpos.app/download