How to track restaurant no-shows properly

In Resos booking data from August 2025 to July 2026, 27% (656 of 2,417 locations) recorded zero no-shows. Some of those may genuinely have had none, but a recorded zero can also mean the no-show status was never used. The source is the Resos No-Show Index (Resos booking data, Aug 2025–Jul 2026).

A rate is only as good as the marking behind it. This page is the routine: what to mark, what to keep separate, and how to calculate a figure you can compare with the Index.

Why recorded no-shows undercount

Across all restaurants in the Resos No-Show Index, the recorded no-show rate was 2.33%. Among the 1,761 restaurants that actively recorded no-shows, it was 2.49%. 72.9% of eligible locations recorded at least one no-show. This suggests that inconsistent marking pulls recorded rates down.

The 27% line and the 72.9% line are the two published ways of stating that coverage. A recorded zero can be genuine, or it can mean the no-show status was never used. The data cannot tell those apart.

No-show vs late cancellation vs walk-in

Resos counts a no-show only when staff or an integration mark the reservation as a no-show. Cancellations, declined booking requests and walk-ins are not part of the rate.

A walk-in cannot fail to arrive for an advance reservation, so walk-ins stay out of the denominator. A cancellation is the guest releasing the table. It is a different event, even when it arrives late.

In the Resos data, 56% of cancellations arrived within 24 hours of service or after the booking had started (317,541 of 566,049 cancellations). Track late cancellations separately from no-shows. The policy side of that split is a restaurant cancellation policy.

A simple marking routine for every service

1. At the end of service, walk the book and mark every advance reservation that did not arrive.
2. Agree who marks. One person closing the service, using the same rule every night, beats three people guessing.
3. Agree what happens when a guest arrives after the table was marked. Pick a grace period the team will actually follow, then mark only after that period has passed. If the guest arrives, change the status.
4. If an integration writes the status, check that it uses the same meaning as the team: a marked no-show, not a cancellation and not a walk-in.

The routine has no published minute count. The point is that the same person, or the same integration, applies the same rule every service.

How to calculate your no-show rate

Recorded no-shows ÷ (honoured reservations + recorded no-shows), for advance reservations only.

Leave out walk-ins, cancellations, declined booking requests and waitlist entries.

The worked example is the Index itself, not a fictional venue. Honoured reservations plus recorded no-shows is the denominator: 3,768,761 mature advance reservations. The numerator is 87,953 recorded no-shows. That recorded no-show rate is 2.33%.

Country, party size, lead time, lunch and dinner sit on what is a normal restaurant no-show rate and on the Index. Use those when you compare, so a Tuesday lunch is not judged against a Saturday dinner in another country.

What to compare your rate against

Compare a marked rate with your own country and with lunch or dinner, not only with the headline. Both cuts are published on the explainer and on the Index. Restaurant overbooking uses the same marking-coverage figure, so the two pages stay on the same definition.

If your rate is zero, go back to the routine before you treat zero as a result.

Using your no-show history

Once the status is real, the history is usable. Repeat no-show guest alerts show that history on a later booking. A no-show fee is a separate policy for charging a missed reservation. Table booking reminders are a separate message before service. What the booking data shows about a delivered SMS, as an association only, is in SMS reminders and restaurant no-shows.

None of those links is a measured effect. The Index does not say what happens when a restaurant switches a tool on.

Tracking no-shows in Resos

In Resos, a no-show is a booking status. Staff set the booking to No-show. Resos also flags the guest when a booking is set to that status. Changing the booking status back later does not remove the flag.

Flagging a no-show guest works on every plan, the Free plan included. A no-show fee does not: it needs an active paid plan and the Online payment add-on. The two tools are written up in the no-show add-ons.

To turn flagging on, go to Menu → Settings & add-ons → Bookings → Flag guests. You can flag from the booking on Menu → Calendar, or from the guest under Menu → Customers. A flagged guest can still book. The flag is a warning for your own team, not a block.

Method and caveats

• Same base as the Index: 2,417 currently active Resos restaurants, 3,768,761 mature advance reservations, 1 Aug 2025 to 31 Jul 2026, extracted 17 Aug 2026. Cite the Resos No-Show Index (Resos booking data, Aug 2025–Jul 2026).
• 27% (656 of 2,417 locations) recorded zero no-shows. A recorded zero may be real, or it may be non-use of the no-show status. The data cannot tell them apart.
• Among the 1,761 restaurants that recorded at least one no-show, the recorded no-show rate was 2.49%.
• Rates are booking-weighted and staff-recorded. The sample is not random and has survivor bias.

Frequently asked questions

How do you calculate a restaurant no-show rate?

Recorded no-shows ÷ (honoured reservations + recorded no-shows), for advance reservations only. Exclude walk-ins, cancellations, declined booking requests and waitlist entries. On the Index totals, 87,953 recorded no-shows out of 3,768,761 mature advance reservations is a recorded no-show rate of 2.33%.

Should late cancellations count as no-shows?

No. Track late cancellations separately from no-shows. In the Resos data, 56% of cancellations arrived within 24 hours of service or after the booking had started.

Why is my no-show rate zero?

Check whether the team marks no-shows. In Resos booking data from August 2025 to July 2026, 27% (656 of 2,417 locations) recorded zero no-shows. Some of those may genuinely have had none, but a recorded zero can also mean the no-show status was never used.

When should staff mark a no-show?

After the grace period the team has agreed, and again in an end-of-service check so the book is complete. The dataset does not name a minute count.

What is a normal no-show rate?

The recorded no-show rate across the Resos dataset was 2.33%. The comparison tables are in what is a normal restaurant no-show rate and on the Resos No-Show Index.

Related reading

How to reduce no-shows in the restaurant industry is the practical companion. Restaurant overbooking uses the same recorded-rate definition. The academy glossary and the academy note on no-show rate are links only, not the source of these figures: no-show and no-show rate.