What a no show costs you on a Saturday night

Ask most operators what a no show costs and they will give you a feeling rather than a figure. It is frustrating, it is unfair to the team, it wastes the prep. All true, and none of it helps decide what to do. Whether to send a second reminder, whether to ask for deposits, whether to hold tables for walk-ins: each of those decisions is a trade, and you cannot weigh a trade without knowing what is on the other side of the scale.

The surprise, when you do the sum properly, is how much the answer depends on when the no show happens. An empty table for two at nine o'clock on a fully booked Saturday can cost several times what the same table costs at noon on a quiet Tuesday, even though the diary shows exactly the same thing: a booking that never arrived. The difference is whether you could have sold that table to someone else.

This guide shows how to work out what a no show costs in your own venue, with your own numbers. It covers average spend per head, the difference between lost revenue and lost margin, the costs that have already been spent by the time the table is empty, the second-seating effect on a busy night, and how to turn a per-table cost into a monthly figure that tells you how hard to push on each service.

Most attempts to cost a no show take the average bill and stop there. That gives you a number, but it is the wrong one for most decisions, because it assumes two things that are rarely true at the same time: that the guests would have spent the average, and that nobody else could have used the table.

On a quiet night, the second assumption is wrong in your favour. If the room is half empty, an absent party does not displace anyone. The table would have sat empty anyway, so the loss is the margin those guests would have produced, minus nothing, plus whatever you spent preparing for them.

On a full night, the second assumption is wrong against you. If you turned guests away because the diary said the room was full, the absent party did displace someone, and possibly more than one party if the table could have been turned twice. The loss is not only their meal but the meals of the guests you sent elsewhere.

So the useful answer is not a single figure but a figure per service, calculated with a few simple sums. The rest of this guide walks through them.

Every sum below starts from average spend per head: the average amount each guest spends, before tax, on a given service. It is not the same as the average bill, which depends on party size, and it is worth calculating separately for lunch and dinner, for weekdays and weekends, because the figures are often very different.

To calculate it, take total sales for a service, before tax and after discounts, and divide by the number of covers served. If Saturday dinners over the last month took 42,000 across 1,000 covers, average spend per head on Saturday dinner is 42. If Tuesday lunches took 6,000 across 250 covers, it is 24.

Use spend per head from the same service you are costing. A Saturday no show should be costed at Saturday's spend, not the week's average. Our guide to increasing average spend covers how to raise the figure itself, which also raises what every no show costs.

The simplest sum is lost revenue: the spend per head multiplied by the number of guests who did not arrive.

A table for two that no shows on Saturday dinner, at 42 a head, is 84 of lost revenue. A table for two at Tuesday lunch, at 24 a head, is 48. A table for six on Saturday is 252.

Lost revenue is a useful headline, but on its own it overstates the loss on quiet services and understates it on busy ones. It overstates it on quiet services because revenue is not profit: some of that 48 would have been spent on the food the guests ate. It understates it on busy ones because it ignores the guests who could have filled the table. The next two sums correct both.

Revenue is what the guests would have paid. Margin is what you would have kept after the direct cost of serving them, mainly the food and drink they would have consumed. This is the idea behind contribution margin: the amount each sale leaves over to pay for everything else.

If your food and drink cost on Saturday dinner averages around 30 percent of sales, then of that 84 of lost revenue, roughly 25 would have been spent on ingredients that you did not use. The lost margin is closer to 59. That sounds better, until you account for the costs that were spent before the guests failed to arrive.

By the time a table is empty at nine o'clock, much of the cost of serving it has already been paid. The kitchen prepped for a full room, and some of that prep will not keep. The rota was built for the booked covers, so labour for those guests was paid regardless. Some ingredients were bought specifically for the night. These sunk costs do not change the lost margin, but they explain why a no show feels worse than a missed sale: you paid to be ready for guests who never came.

The largest part of what a no show costs on a busy night is usually the third sum: the guests you could have seated instead.

If the diary was full and you turned away walk-ins or declined bookings, then an empty table is not just a lost meal. It is a meal you could have sold to someone else. If that table could have been turned twice in the evening, and the no show booking held it for the first sitting, you may have lost the first sitting entirely and seated the second late, or not at all.

This is where table turnover rate matters. A table that turns twice on a Saturday represents two sets of guests. A no show that empties the first sitting, without warning, often costs the whole first turn, because by the time the hold window has passed it is too late to fill it with anyone but a walk-in who happens to be at the door. Our guide to table turnover rate covers how turns work on a busy night.

Here is an illustrative comparison for a restaurant with two sittings on Saturday night and a quiet Tuesday lunch. The figures are examples to show the sums, not typical results.

Saturday dinner:

The direct lost revenue is 4 × 42 = 168. The table could have been sold to one of the walk-in parties turned away earlier, so the opportunity cost is real: roughly the same 168 that could have been taken from them. The late walk-in recovers part of it, but a shortened sitting usually spends less. A reasonable estimate of what the no show costs that night is somewhere between 120 and 168 in revenue, plus the stress of a disrupted service.

Tuesday lunch:

The lost revenue is 4 × 24 = 96, but nobody else could have used the table, so there is no opportunity cost. The lost margin, at around 30 percent food cost, is around 67, and some of the prep may be used at dinner. The real cost is smaller and mostly the margin those guests would have produced.

The two empty tables look identical in the diary. Their costs are not.

A per-table cost becomes useful when you multiply it across a month. Take each service in turn: the number of no show covers, the average spend per head for that service, and a judgement about whether those tables could have been resold.

For example, suppose Saturday dinners in a month produced 40 no show covers at 42 a head, on nights when the diary was full. That is 1,680 of lost revenue, most of which could have been resold. Tuesday lunches produced 20 no show covers at 24 a head, on quiet days: 480 of lost revenue, little of which could have been resold. Even though Tuesday's no show rate might be similar, Saturday's problem is worth far more effort.

This is why your no show rate should always be read alongside the value of each service. A high rate on a quiet service may cost less than a low rate on a full one.

Once you know what a no show costs on each service, the response becomes proportionate. Services where no shows are cheap need only the basics: verified contact details, a clear confirmation and a reminder. Services where they are expensive justify more: a second reminder on the day, a live waitlist, a few tables held for walk-ins, and, where demand is strong, a card hold or deposit for large parties.

The sum also tells you how much a measure is worth. If no shows on Saturday dinners cost around 1,600 a month, a measure that halves them is worth around 800 a month. If a deposit would also lose some genuine bookings, you can estimate whether the saving outweighs the loss. Our guides on how to reduce no shows and reservation deposits set out the options in order of friction.

Large parties change the arithmetic in two ways. A no show from a party of ten empties ten seats at once, often across combined tables that could have seated two or three smaller parties. And large parties often book further ahead, when a no show has more time to develop. That is why many venues treat large parties differently, with a confirmation a few days ahead and sometimes a deposit, while leaving small bookings free.

When you cost no shows, split them by party size as well as by service. If a small number of large-party no shows account for a large share of the cost, the fix can be targeted at them alone.

Some costs of a no show never appear in any of these sums, but they are real.

These are hard to put a number on, which is exactly why they are worth mentioning. A no show problem that seems modest in pure revenue can still be worth fixing for these reasons.

A live waitlist reduces what a no show costs on a busy night, because it gives you someone to call the moment a table falls empty. If a cancelled or no show table can be filled within fifteen minutes from the waitlist, the opportunity cost largely disappears. That is why a waitlist and good reminders often do more to reduce the cost of no shows than any deposit, because they recover the table rather than merely discouraging the absence. Our guide to waitlist management covers running one that guests trust.

Once the sums are set up, they take minutes. A short weekly review can look at no show covers by service, the spend per head for each, and whether those nights were full. A monthly total, by service, tells you which nights are worth more effort. Keep the method the same each time, so the trend is meaningful. When you change something, such as adding a reminder or a waitlist, the monthly cost should fall on the services you targeted.

The people taking bookings and running the door affect no shows more than anyone, and they respond to numbers they can understand. Telling a host that "an empty four-top on Saturday at seven costs us about as much as a quiet Tuesday lunch for the whole room" makes the stakes concrete in a way a percentage never does. It also explains why the team should read back contact details on every phone booking, keep the waitlist alive and call the next party the moment a table frees up.

Share the monthly figure too, by service, and the change after each improvement. When the team can see that a new reminder or a better-run waitlist reduced the cost of no shows on the busiest nights, they keep doing the things that caused it.

If you operate more than one venue, cost no shows for each venue and each service separately before adding them together. A group total can hide a single venue with a serious weekend problem behind several venues that are running well. Different venues also have different spend per head and different turnover, so the same number of empty covers can cost very different amounts in each. Compare the trend for each venue against itself first, and only then compare venues with each other.

The sums in this guide need three numbers: how many covers did not show, on which service, and what guests spend there. In servearound, the reservations report in Analytics shows covers, bookings and the no show, cancellation and seated rates by meal period, source and party size, set against the period before. The payments report shows the average and median check, and the table report shows revenue per seated hour and the minutes each table went unsold, which is the clearest picture of what an empty table on a busy night really cost.

On the booking side, the reservations module verifies guests' email addresses, sends a reminder as many hours ahead as you choose, and gives guests a link to change or cancel, so more empty tables arrive as cancellations you can fill from the waitlist.

Yes, if you are costing the whole table. Drinks are part of what the guests would have spent, and on many services they are a large share of it.

Then the cost is much smaller: the difference between what the booked party would have spent and what the walk-in spent, plus any disruption. Record it, because it shows how well your walk-in handling recovers the loss.

Briefly, yes, if only to confirm that they are cheap. It stops you applying expensive measures, such as deposits, where they are not needed.

Not very. The aim is to know roughly how much each service's no shows cost, to the nearest few hundred a month, so you can decide where to act.

Usually much more, because food is often prepared for each guest in advance and the room may be small. That is why prepaid bookings are common for tasting menus.

Usually much less, because the guests still eat and pay. A very late arrival on a busy night can still cost something, though, if it delays the second sitting at that table. Note late arrivals separately so you can see whether they are a problem on particular services.

Labour scheduled for the booked covers is usually paid regardless, so it is a sunk cost rather than part of the lost margin. Over time, though, regular no shows push the rota higher than trade justifies, which is a real cost worth noting.

What a no show costs depends on when it happens. Start with average spend per head for the specific service. Calculate lost revenue, then lost margin, and remember the costs already spent on prep and staffing. On full nights, add the cost of the guests you turned away, which is usually the largest part.

Multiply across a month by service and party size, and you will see exactly which nights deserve stronger measures. A Saturday no show and a Tuesday no show look the same in the diary, and the sum is the only way to tell them apart.