Empty Tables, Real Numbers: What Australian No-Show Data Says About Deposits and Reminders

Most booking policies are set on memory of the worst Saturday. Recorded reservation data tells a calmer story about restaurant no-shows, and a more useful one about late cancellations, reminders and where deposits belong.

Every operator has a no-show story: the table of eight that never arrived on the busiest night of the month. Those stories shape booking policy more than data does, and the result is often a deposit on every booking or none at all. Recorded data on restaurant no-shows suggests a more targeted approach. The headline rate is lower than many operators assume, the bigger leak is late cancellation, and the tools that help are not the same on a Tuesday lunch as on a Saturday dinner.

What the recorded data shows

The most detailed recent dataset we found comes from a booking platform. The Resos No-Show Index 2026 covers 3,768,761 reservations across 2,417 restaurants between August 2025 and July 2026, and records an average no-show rate of 2.33 per cent. Those 87,953 recorded no-shows represent 283,728 booked covers that never arrived. The Australian subset is lower: 1.41 per cent, across 251,804 reservations at 213 Australian restaurants, a figure also reported by Skai Solutions citing the same index.

These are vendor figures, drawn from one platform’s customers and limited to no-shows that venues recorded. We did not find an independent Australian dataset of comparable size, so the numbers are best read as a benchmark to compare against your own records rather than as an industry statistic.

Dinner, lunch and the late cancellation

The pattern within the data matters more than the average. In the same index, the no-show rate at dinner (2.63 per cent) runs 43 per cent higher than at lunch (1.84 per cent). A blanket policy that treats both services the same will either over-protect lunch or under-protect dinner.

Late cancellations are the larger issue. Of 566,049 cancellations in the dataset, 317,541, or 56 per cent, came within 24 hours of the booking or after it had started. For a venue, a table cancelled at 4 pm for 7 pm is often as hard to resell as one that simply does not appear, and it does not show up in a no-show count at all.

The empty table you remember is a no-show. The empty table that costs you most is usually a cancellation that arrived too late to resell.

What reminders and deposits are associated with

Reminders are the cheapest tool and the data is encouraging, with an important caveat. Reservations with a delivered SMS reminder showed a 16 per cent lower recorded no-show rate than reservations at the same restaurants without one. Resos is careful to say that “everything in this section is an observed association, not a measured effect”, because no bookings were randomly assigned.

Deposits show a larger gap, with a larger caveat. Reservations requiring no-show protection recorded a 0.80 per cent no-show rate, against 2.47 per cent for reservations without a payment requirement. Resos publishes these “as raw descriptive rates only”, noting that restaurants that require deposits differ from those that do not. It adds that a matched study of restaurants switching payment features on produced estimates with uncertainty intervals “wide enough to include no change in either direction”. In plain terms, deposits sit alongside lower no-show rates, but the data does not prove that deposits cause them.

Matching the policy to the booking

That points to a tiered policy rather than a single rule. High-value, hard-to-resell bookings, such as large groups, set menus, degustations, peak Saturday sittings and special event nights, carry the most risk and justify the most protection. Small weekday lunch bookings carry the least, and a deposit there may cost more in lost bookings and guest goodwill than it saves.

The same logic applies to cancellation windows. If more than half of cancellations arrive inside 24 hours, a reminder that asks guests to confirm or release the table 24 to 48 hours ahead gives the venue time to fill the gap from a waitlist. How deposits and cancellation fees are worded, disclosed and refunded also matters, so take advice on how your terms apply under the Australian Consumer Law before introducing or changing them.

A booking-policy reset in six steps

1. Measure your own rate by service. Pull 12 months of booking data and calculate no-shows and same-day cancellations separately for lunch, dinner, weekdays and weekends.
2. Price the empty seat. Multiply average spend per head by the covers lost each month, so the policy is judged against a real figure rather than a memory.
3. Send reminders by default. Make a confirmation request 24 to 48 hours before every booking standard, with a simple way to cancel or change.
4. Protect only the bookings that need it. Reserve deposits or card guarantees for large groups, set menus, event nights and peak sittings where the data shows the highest risk.
5. Build a waitlist that can absorb late cancellations. A released table is only recovered if there is someone ready to take it.
6. Review quarterly. Compare no-show and cancellation rates before and after each change, and adjust the tiers rather than adding blanket rules.

Booking policy is one part of a wider forecasting discipline: the covers you expect, the covers that arrive and the staff you roster for them. We explored that link in Your Sales History Is a Forecasting Tool. At Alto Cibum, guest-journey design, booking strategy and covers forecasting are part of our food and beverage consulting work, and the starting point is always the venue’s own data rather than the worst night anyone can remember.

Robert Clifford
Managing Director, Alto Cibum