No-shows, deposits and cancellations

Measuring your no-show rate properly

How to calculate a no-show rate you can act on: clear definitions, booking and cover formulas, splits by day and party size, and a monthly dashboard.

Ava research teamPublished 6 min read

On this page
  1. Why most no-show rates are wrong
  2. Define each outcome once
  3. The formulas
  4. Bookings or covers?
  5. Split the numbers before you judge them
  6. Benchmark by day against your own history
  7. Measurement traps to avoid
  8. Keep the data clean
  9. Share the numbers with the team
  10. A monthly dashboard of six numbers
  11. Your first month: what to expect
  12. Turn the numbers into one decision
  13. Frequently asked questions
  14. References

Key takeaways

  • Separate no-shows from late cancellations, partial no-shows and venue cancellations; lumping them together hides the fix.
  • Track two rates: bookings that failed and covers that failed. Large parties can make them tell different stories.
  • Benchmark each day of the week against its own history, not against other venues or the busiest night.
  • Review six numbers monthly and change one thing at a time so you can see what worked.

Why most no-show rates are wrong

Ask three managers for their no-show rate and you will get three answers built three different ways. One counts late cancellations, one counts covers, and one remembers last Saturday. None is wrong exactly, but none can tell you whether reminders are working or whether a deposit on Fridays is worth the friction.

A useful rate starts with fixed definitions, applied the same way by everyone who closes a booking in your system. Get those right and the arithmetic is easy.

Define each outcome once

Write these definitions into your booking procedures and use the same status codes in the system.
OutcomeDefinition to use
SeatedThe party arrived and was seated, even if late
No-showNobody arrived and there was no contact by 15 minutes after the booking time
Late cancellationThe guest cancelled inside your notice period, for example within 24 hours
Early cancellationThe guest cancelled with more notice than your policy asks for
Partial no-showThe party arrived with fewer guests than booked after the numbers deadline
Venue cancellationYou cancelled or moved the booking
Walk-outThe party arrived and left before ordering, for example after a long wait

Two rules make the definitions stick. First, every booking gets exactly one outcome by the end of the night. Second, nobody deletes a booking that failed; it gets a status instead. Deleted bookings are the most common reason a no-show rate looks better than it is.

The formulas

  1. No-show rate (bookings) = no-shows ÷ bookings due that day, excluding early cancellations and venue cancellations.
  2. No-show rate (covers) = covers booked by no-show parties ÷ covers booked for bookings due.
  3. Late cancellation rate = late cancellations ÷ bookings due.
  4. Partial no-show rate = missing covers in seated parties ÷ covers booked for seated parties.
  5. Refill rate = late cancellations and no-shows whose table was resold ÷ all late cancellations and no-shows.
  6. Lost covers = covers from no-shows, late cancellations and partial no-shows, minus covers you refilled.

Lost covers is the number that matters most for the business, because it combines the problem and what you did about it. The others tell you where the problem comes from.

Bookings or covers?

Track both. A booking-based rate treats a couple and a party of twelve the same, which is fair for measuring guest behaviour but misleading about money. A cover-based rate weights big parties properly but can swing wildly on one bad group.

Example: on a Saturday with 40 bookings and 140 covers, two couples and one party of ten fail to arrive. The booking rate is 3 out of 40, or 7.5%. The cover rate is 14 out of 140, or 10%. The gap tells you large parties are the bigger problem that night, which points to a different fix from reminder texts for couples.

Split the numbers before you judge them

  • Day of the week and service: Tuesday lunch and Saturday dinner are different businesses.
  • Party size: one to two, three to five, six to nine, ten or more.
  • Lead time: booked today, within a week, within a month, more than a month ahead.
  • Channel: your website, a booking platform, phone, social media messages.
  • Protection: no card, card hold, deposit, prepaid.
  • Guest history: first booking or returning guest.

Splits are where the decisions come from. If no-shows cluster in bookings made more than a month ahead through one platform, you need reminders for that group, not a deposit on every table.

Benchmark by day against your own history

There is no reliable public benchmark for restaurant no-show rates, so the most useful comparison is your own history: this October's Saturdays against last October's, not against a figure from a supplier's blog. Your trading pattern moves with the season and the market, and so will your rate.

Market conditions also vary sharply by format. In September 2026, like-for-like sales at managed pub groups grew 3.8% while managed bar groups fell 1.7%, and managed restaurant groups grew 2.6%[1]. Demand also shifts month to month: 30.7% of trading accommodation and food businesses said domestic demand fell in September compared with August[2].

Like-for-like sales growth by format, managed groups, September 2026Year on year. Managed bar groups fell 1.7% in the same month; benchmark against your own history.
Like-for-like sales growth by format, managed groups, September 2026Column chart. Pubs: 3.8%; All managed groups: 2.7%; Restaurants: 2.6%.0%1%2%3%4%Pubs: 3.8%3.8%PubsAll managed groups: 2.7%2.7%All managed groupsRestaurants: 2.6%2.6%RestaurantsLike-for-like sales growth by format, managed groups, September 2026Column chart. Pubs: 3.8%; All managed groups: 2.7%; Restaurants: 2.6%.0%1%2%3%4%Pubs: 3.8%3.8%PubsAll managed groups: 2.7%2.7%All managedgroupsRestaurants: 2.6%2.6%Restaurants

Source: NIQ (CGA by NIQ) with RSM, 2026[1]

Show the data
ItemValue
Pubs3.8%
All managed groups2.7%
Restaurants2.6%

Christmas and New Year, the sector's crucial trading period[3], deserve their own baseline. Compare December with December, and set peak-date measures against peak-date history.

Measurement traps to avoid

  • Counting walk-ins in the denominator. They were never bookings, so they make the rate look lower than it is.
  • Treating a moved booking as a cancellation and a new booking. Record it as moved, or the rate double-counts.
  • Measuring only on the nights that felt bad. Use every booking, every night, or the figure becomes a story.
  • Comparing a month with a bank holiday against one without. Note the calendar differences when you compare.
  • Ignoring bookings made under staff names for regulars or VIPs. They fail too, and they are easy to forget.
  • Changing the hold time mid-month. A 15-minute rule one week and 30 minutes the next will move the rate on its own.

Most of these traps make the rate look better than reality, which is why they survive. A slightly worse but honest number is far more useful than a flattering one.

Keep the data clean

  1. Close every booking every night

    The closing manager gives each booking an outcome before leaving. Unclosed bookings count as unknown, not as seated.

  2. Use the 15-minute rule

    Mark a no-show only after the hold time in your policy, and note any contact the guest made.

  3. Timestamp cancellations

    Record when the cancellation arrived, so late and early cancellations are sorted correctly.

  4. Record party size on arrival

    Note the number seated, so partial no-shows are visible.

  5. Flag venue errors

    If the booking failed because of your mistake, mark it as a venue issue, not a no-show.

Share the numbers with the team

Front-of-house staff influence the rate more than anyone: they take phone bookings, send or forget reminders, close bookings at night and greet the parties that arrive short. Show them the monthly figures, briefly and without blame, and explain what each one means. A team that knows the Saturday no-show rate fell after they started calling large parties will keep calling them.

Keep the conversation on process, not on individual guests. The aim is to find the gaps in how bookings are taken and confirmed, not to build a list of people to complain about.

A monthly dashboard of six numbers

  1. No-show rate by bookings, by day of the week.
  2. No-show rate by covers, overall.
  3. Late cancellation rate.
  4. Partial no-show rate for parties of six or more.
  5. Refill rate.
  6. Lost covers, compared with the same month last year.

Six numbers fit on one page and can be read in five minutes at a management meeting. More than that and nobody looks.

Your first month: what to expect

The first month of clean data usually brings a surprise. Rates are often higher than managers expected, because bookings that used to be deleted are now counted, and the split by lead time or channel often points somewhere nobody was looking. Resist the urge to act on everything at once. Pick the biggest single source of lost covers and fix that first.

Example: a pub's first clean month shows a 4% booking-based rate overall but 11% for Sunday lunch tables booked more than three weeks ahead. The manager adds a reminder on Friday for those bookings only, leaves everything else alone, and checks the same figure four Sundays later.

Turn the numbers into one decision

Use each month's review to make at most one change, such as adding a reminder for bookings made more than three weeks ahead, and then hold it for at least a month. Changing three things at once leaves you guessing which one worked. Evening service carries most of the risk, because that is where most eating-out spend happens: restaurants, cafes, bars and nightclubs took 42% of time-stamped spending between 6pm and 6am in an ONS sample week[4].

For the tools to act on what you find, see refilling late cancellations, partial no-shows and whether deposits put guests off.

Frequently asked questions

How do you calculate a restaurant no-show rate?

Divide the number of bookings where nobody arrived and nobody made contact by the number of bookings due that day, leaving out early cancellations and bookings the venue cancelled. Track a cover-based rate alongside it.

Should late cancellations count as no-shows?

No. Track them separately. They need different fixes: a late cancellation gives you a chance to refill the table, a no-show does not.

What is a good no-show rate for a restaurant?

There is no reliable public benchmark. Compare each day and season with your own history and aim to bring lost covers down month by month.

How often should we review no-show figures?

Monthly for decisions, with a quick weekly glance at the busiest nights. Make one change at a time and hold it long enough to see its effect.

References

  1. NIQ (CGA by NIQ) with RSM (2026). Bright September gives hospitality groups their best growth of 2026 (NIQ RSM Hospitality Business Tracker) (opens in a new tab) ↑

  2. Office for National Statistics (2026). Business insights and impact on the UK economy, wave 166 dataset (opens in a new tab) ↑

  3. NIQ (CGA by NIQ) (2026). Hospitality loses 19 licensed premises a week ahead of crunch Christmas (Hospitality Market Monitor Q3 2026) (opens in a new tab) ↑

  4. Office for National Statistics (2022). The night-time economy, UK: 2022 (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo