How it works

What is table turnover in a restaurant?

Table turnover is how many times you sell the same seat in one service, and Skai Solutions calculates it as covers divided by seats, so a 60 seat dining room that serves 90 people at dinner has turned 1.5 times.

It is the one operating number that does not appear anywhere in your accounts, which is why it gets either ignored or chased far past the point where it helps. This page sets out how the figure is built, what a real service does to it, the three things it hides, and the kind of venue that should leave it alone entirely.

Book a Free Demo

How do you calculate table turnover?

Covers divided by seats, for one service period, in one dining room. Not per day, and not across the whole week, because a number averaged over a dead Tuesday lunch and a full Saturday dinner describes neither.

1. Count seats, not tables. Seats are what you sell. Two four tops pushed together are eight seats whether they are billed as one table or two.
2. Count covers from the POS, not the book. Covers are people served. The booking diary records intentions, and the gap between the two is the whole subject of the second half of this page.
3. Fix the service window before you start. Dinner 5pm-10pm is one service. If you decide mid-month that dinner now starts at 5.30pm, every comparison you make afterwards is against a different denominator.
4. Divide. 90 covers across 60 seats is 1.5. 140 covers across 60 seats is 2.3.

Some venues run it per table instead, dividing covers by the number of tables. That version moves whenever you reconfigure the floor, which makes it useless for comparing one month to the next. Seats are stable, so seats win.

What is a good table turnover rate?

There is no published Australian or New Zealand benchmark, and this page is not going to invent one. No government statistical agency collects it, and the figures that circulate online are almost all American casual dining averages repeated between vendor blogs with no original measurement underneath them. What can be said honestly is structural rather than numerical.

• Turnover falls as average spend rises. A tasting menu sells a seat once a night by design, and that is a business model rather than a failure.
• It falls as the party size rises, because a table of ten occupies a footprint for longer and rarely leaves together in the way two people do.
• It is capped by your kitchen, not your dining room. A pass that can plate 70 mains an hour sets your ceiling however many seats you own.

The useful comparison is your own venue against itself, same service, same season, last year. Compare that to what the average restaurant makes in Australia and you can see whether a turnover change actually reached the bank account or just moved the mix.

What table turnover hides

Three things, and the first one is the reason most turnover projects disappoint.

• It counts served covers, not sold ones. A no-show does not reduce turnover, because nobody sat down, but it does hold a table empty through the best part of service. The Resos No-Show Index 2026, built from 3,768,761 real reservations at 2,417 restaurants, recorded 1.41% of Australian bookings as no-shows across 251,804 Australian reservations. The average restaurant in that dataset lost about 36 bookings and about 117 booked covers over the year. Turnover is silent on all of it.
• It says nothing about spend. Two turns at A$38 a head beats one and a half at A$45, but one and a half at A$70 beats both. Turnover without average spend beside it is half a fraction.
• It says nothing about labour. ABS Australian Industry figures released on 18 September 2026 put total labour costs at 30.4% of sales across Australian food and beverage services in 2024-25, up from 30.1%. A second turn bought by adding a floor staff member for the whole service can cost more than it brings, and turnover will still read as an improvement. Our page on restaurant labour cost percentage has the detail.

Book a Free Demo

How do venues actually increase table turnover?

Cheapest first, which is also the order in which the gains are most likely to be real.

1. Stop losing the second sitting to a first sitting that never arrived. A held table is the single largest avoidable loss on most floors. Resos measured 56% of cancellations arriving within 24 hours of the booking or after the start time, and 7.1% landing after the booking had already begun, so the release comes too late to resell. Our page on restaurant no-shows covers what actually reduces them.
2. Shorten the tail, not the meal. Most of the time between turns is not eating. It is the wait for the bill, the wait for the card machine and the wait for someone to clear and reset. Paying at the table takes minutes out of every turn and nobody feels rushed.
3. Match the floor plan to the bookings you actually take. If two thirds of your bookings are pairs and two thirds of your seats are in six tops, you lose seats every service to arithmetic rather than to demand.
4. Stagger the book. Three sittings at 6pm, 8pm and 9.30pm read well in a diary and arrive as one wall of tickets. Fifteen minute offsets spread the same covers across a kitchen that can cope with them.
5. Then, and only then, look at time limits. A two hour limit on a table is the most visible lever and the one guests resent most.

Turnover only matters once the room is full, so the prior question is how many people arrive at all. Our page on how many customers a cafe gets a day builds that count from published New Zealand figures.

Turnover only matters against the number of people who actually arrive. Our page on how many customers a cafe gets a day covers that side of it.

Who should not chase table turnover?

Plenty of venues, and being honest about that is the point of this section.

• Fine dining and degustation. The business is built on one seating. Turning it twice would mean being a different restaurant.
• Any venue that is not full. If you are running at 60% occupancy, turnover is not your constraint and speeding up service will not create demand. Fill the room first.
• Cafes and daytime venues where the number is already high. A cafe can turn a table five or six times over a morning without anyone managing it. Our page on how many staff a cafe needs is a more useful place to look.
• Takeaway-weighted venues. If half your revenue never sits down, seats are not the denominator that describes your business. Phone ordering for takeaways is the relevant page.

Does this apply in New Zealand?

The arithmetic does, the measured no-show data does not. New Zealand did not meet the publication threshold in the Resos index, so there is no New Zealand no-show rate in it and nobody should imply one. On the cost side, the Restaurant Association of New Zealand's 2025 Hospitality Report puts wage costs at about 40% of hospitality revenue, on a wider definition than the ABS uses, which is one reason a New Zealand operator should not read the Australian labour share as their own. Our page on hospitality wages in New Zealand sets out the rates that drive it.

Where the phone sits in this

On one specific part of it. The turn you lose to an empty held table is decided before service starts, in the booking and in the cancellation that either reaches you in time to resell the table or does not. A call that rings out at 6.40pm because every hand is carrying plates is a cancellation you find out about at 7.15pm, with the table gone for the night.

That is the honest size of the claim. Nothing about answering a phone shortens a meal or clears a table faster. If your phone mostly carries enquiries and opening hours, a basic message-taking plan is the right buy and it is not us. Where an AI receptionist earns its place is a venue whose bookings and cancellations arrive by phone during service, when the floor cannot answer. The cheapest way to answer a restaurant phone prices that honestly against every alternative, free ones included.

Turning tables faster only helps if the bill holds up. Average spend per head shows why the published New Zealand data has average basket size nearly flat for nine years while transaction counts rose.

Turns multiply seats, so the seat count is the other half of this number. Restaurant seating capacity per square metre sets out the building code figure and the permit condition that usually sits below it.

Cafes turn faster than restaurants and answer fewer of their calls because of it. An AI receptionist for a coffee shop covers what happens to the phone during a morning rush and whether it is worth covering.

Questions operators ask about table turnover