To compare no-shows across restaurant shifts, divide the number of no-show parties by the number of parties with a final seated or no-show outcome. Exclude canceled parties and unresolved bookings from that denominator. Use the same definitions and cutoff for every shift.
This gives you an operational rate: of the parties expected to arrive whose outcomes are known, how many did not show? It is different from a no-show cost estimate and from a cancellation rate. Keeping those measures separate makes it easier to see which part of the booking process needs attention.
Define the outcome of each booking
Start with one record per guest party, not one record per physical table. A party of six seated across two tables is still one party. If a group changes its reservation time or size, agree on how staff will identify the original booking so it is not accidentally counted twice.
For this calculation, assign each booking to one of four outcomes:
- Seated: The party arrived and was seated, even if fewer guests came than originally booked. Record the actual party size separately when possible.
- Canceled: The reservation was canceled before the restaurant classified it as a no-show. A cancellation may be early or late; keep that timing as a separate detail.
- No-show: The party did not arrive within the restaurant's stated grace period and the booking was not canceled.
- Unresolved: Staff cannot yet establish the outcome, or the booking has not been updated.
Choose a grace period that fits your service and apply it consistently. A party that arrives late but is seated should end up as seated, not remain a no-show because someone marked it absent at the original booking time. Likewise, a guest who calls to cancel should not be treated as absent just because the cancellation was close to service.
Decide what happens to duplicate bookings, rescheduled bookings and walk-ins before running a report. A practical rule is to remove confirmed duplicates, count a rescheduled party against the shift it ultimately booked, and keep walk-ins outside the reservation no-show rate. Document exceptions so a manager can investigate rather than guess.
Choose a denominator that answers the question
The calculation is no-show parties ÷ (seated parties + no-show parties) × 100. The numerator and denominator must cover the same shift, booking channels and reporting period. Canceled bookings are useful information, but they do not belong in this particular denominator because the restaurant knew those parties were not coming.
The following inputs are hypothetical and describe one completed shift. They are party counts, not guest covers or physical tables.
| Outcome | Parties | Count in rate? |
|---|---|---|
| Seated | 40 | Denominator |
| No-show | 5 | Numerator and denominator |
| Canceled | 10 | Neither |
| Unresolved | 5 | Neither, until classified |
| Total booking records | 60 | 45 have a seated or no-show outcome |
The rate is 5 ÷ 45 × 100, or 11.1% when rounded to one decimal place. If the 10 canceled parties were added to the denominator, the result would be 5 ÷ 55 × 100, or 9.1%. That lower figure answers a different question: no-shows as a share of seated, no-show and canceled parties. It can make a shift with more cancellations appear to have a better arrival record.
The five unresolved records also matter. If they are later classified, recalculate the rate rather than treating them as seated by default. If all five turn out to be seated, the rate becomes 5 ÷ 50, or 10.0%. If all five are no-shows, it becomes 10 ÷ 50, or 20.0%. Until staff resolve them, the 11.1% figure is provisional.
Track cancellations separately if you want to understand late changes or demand that disappears before service. Define its denominator and cutoff in the report, too. Do not add the cancellation rate to the no-show rate: the measures may use different populations.
Compare shifts on equal terms
A useful shift report shows the no-show count and eligible party count alongside the percentage. A rate based on a handful of bookings can swing sharply when one party is reclassified. Look at several comparable shifts before changing staffing or booking rules.
Match the shifts you compare. Friday dinner and Tuesday lunch may have different booking patterns, party sizes and lead times. Compare the same service window, include the same booking channels and use the same grace period. If a holiday or private event changes the usual mix, label it instead of quietly folding it into an ordinary week.
Party counts and covers answer different questions. In the hypothetical shift above, suppose the 45 eligible parties represented 100 booked guest covers, including 10 covers in the five no-show parties. The party no-show rate is still 5 ÷ 45, or 11.1%. The booked-cover no-show rate is 10 ÷ 100, or 10.0%.
The first measure tells you how often a reservation party failed to arrive. The second shows the share of booked guests in no-show parties. A single missing large group can have a modest effect on the party rate but a much larger effect on the cover rate. Neither figure tells you how many physical tables sat empty: table combinations, walk-ins and reseating affect that outcome.
Keep booked covers distinct from actual seated covers. A party booked for six that arrives with four is seated for the party-rate calculation; the two missing guests are not automatically a no-show party. If partial arrivals matter to your operation, report the difference between booked and actual covers separately.
Compare booking channels as well as shifts when the data supports it. Direct website reservations and discovery-led marketplace bookings can bring different traffic. OpenTable offers reservations, restaurant operations and a diner discovery network. Fleksa offers direct restaurant reservations and availability-qualified Google reservation paths. Ask each vendor how it identifies the booking channel and whether you can export the fields needed for a consistent report.
Audit classifications before acting
Make the closeout routine simple enough for a host or manager to complete after service. Review unresolved bookings first, then inspect no-shows for late arrivals, duplicate records and cancellations entered against the wrong reservation. Use the reservation name, local date and time, party size and any appropriate notes to find the right record.
If a guest was marked absent in error, change the outcome to seated or canceled as appropriate. Preserve the original booking identity, record the reason for the correction and note who made it and when. Recalculate the affected shift and update any report that used the old result. If you cannot establish what happened, leave the booking unresolved rather than forcing it into a category.
Set a reporting cutoff, such as a manager review after the shift and a final check before the weekly report. A cutoff does not prevent later corrections; it tells readers when the snapshot was taken. Include the number of unresolved bookings with the published rate so managers can judge how much confidence to put in it.
Before adopting a new workflow, run a simulated check using fictional reservations in a safe practice environment. Walk through a late arrival, a cancellation, a duplicate and an unresolved record. Confirm how staff would correct each outcome and produce the counts. The simulation must not send guest messages, take real payments or dispatch a courier.
A live service pilot is different. If you choose to run one, assign a manager to monitor real bookings during a limited service period, follow your normal guest-facing procedures and reconcile the results afterward. Do not assume a practice setup proves that every live integration or staff handoff will work.
Check the reporting workflow before you buy
Bring your definitions to a vendor demonstration rather than asking only for a dashboard. Show the vendor the four outcomes and ask how a host records each one, how a manager corrects an error and whether the report can exclude cancellations and unresolved bookings. Ask to see the underlying counts as well as the percentage.
Also ask which staff roles can edit statuses, what correction history is available, how booking-channel attribution works and which fields you can export. If Google bookings are available, ask how they appear alongside website bookings. For a marketplace, ask about permissions, exports and fees; obtain a written quote for the channels and volume you expect. Check any contract questions with a qualified adviser where needed.
When replacing a system, assign a cutover owner and reconcile booking counts by shift before and after the change. Establish an authorized, supported way to transfer booking identity, local date and time, party size, status and appropriate special requests. Ask how staff will handle differences or bookings that arrive during cutover. A rehearsal reduces uncertainty; it does not guarantee uninterrupted service or that no booking will be missed.
Start with your next shift report: write down the four outcomes, count unresolved records and calculate the rate using seated plus no-show parties. Then compare that workflow with your requirements when evaluating a restaurant reservation system and broader restaurant management software. For a separate look at financial impact, see the guide to measuring no-show costs.
Frequently asked questions
Should canceled reservations count as no-shows?
No. For this operational rate, exclude canceled parties from both the numerator and denominator. Track cancellations separately, including when they occurred, if that helps you manage service.
Should the rate use parties or guest covers?
Use parties to measure how often an eligible reservation fails to arrive. Add a separate booked-cover rate when group size matters. Do not mix party counts and cover counts in the same calculation.
What should I do with an unresolved booking?
Exclude it until staff establish its outcome, and report how many records remain unresolved. Once it is classified, recalculate the affected shift and correct any reports that used the earlier figure.
