What is a cover in a restaurant, exactly?

What is a cover? In a restaurant, a cover is one guest: one person who sits down to eat or drink. A table of four is four covers. A restaurant that served 180 covers last night served 180 guests, whether they sat at 45 tables or 60. It is the most common unit in hospitality, used in bookings, in the kitchen, in the till and in every report the owner reads.

It is also one of the most argued-over words in a restaurant, because different parts of the business count covers differently. The host counts covers booked. The till counts covers on checks. The kitchen counts plates. The accountant divides sales by whatever number they are given. When those numbers disagree, as they often do, the floor and the office end up arguing about whether last Saturday was busy.

This guide explains what a cover is, where the word comes from, the three places it is used differently enough to cause arguments, and how to agree one definition for the whole venue. It covers the numbers built on covers, including the covers report, average spend per head, covers per table and booking capacity, and links to longer guides on each.

A cover is one guest. Not one table, not one booking, not one check: one person. When a venue says it does 200 covers on a Saturday, it means 200 people ate or drank there.

Most venues count children as covers if they order their own food, and do not count staff meals, tastings or deliveries. Each venue should write down its own rules, but the core idea is always the same: one cover, one guest.

The word comes from the place setting laid for each guest: the cutlery, plates, glass and napkin that cover a space at the table. In the hospitality industry, laying a cover meant setting a place, and counting covers meant counting places set and used.

Some languages and countries still use a related word for a charge per guest, such as a cover charge for bread and the place setting. That is a different thing, a fee, but it comes from the same idea: the place laid for one person.

Covers matter because almost everything in a restaurant depends on them. The number of guests decides how many servers and cooks are needed, how much food is prepared, how full the room feels and how much is sold. Sales divided by covers gives average spend per head, one of the most useful numbers in the business.

A venue that tracks sales but not covers cannot tell whether a busy week came from more guests or from guests spending more. Our guide to the covers report covers how to read covers each week.

The word cover is used in three places that count it differently:

Each count is correct for its own purpose. The arguments start when one is compared with another without saying which is which.

In bookings, a cover is a guest on a booking. A booking for six is six covers booked. But not every booked cover arrives. Some guests do not show; some parties shrink at the door; some grow. Walk-ins arrive without any booking at all.

So the booking system has several cover numbers: covers booked, covers seated, covers that did not show and walk-in covers. The covers booked figure is useful for planning; the covers seated figure is what actually happened.

The gap between covers booked and covers seated matters. A Saturday with 200 covers booked and 180 seated had 20 covers that did not arrive, or 10 per cent of the book. That is the no-show and late cancellation cost, and it is invisible if only the booked figure is reported.

Our guide to what is a no-show, and when it counts covers how to count the missing covers fairly.

Walk-ins are covers that never appear in the booking system unless the host adds them. A venue that reports covers from its bookings alone undercounts every night it has walk-ins, sometimes by a large margin. On a busy night with a waitlist, walk-ins can be a fifth of all covers or more.

Record walk-ins at the host stand, so the cover count is complete.

The host stand is the best place to count covers, because it is where every guest arrives. A host who records each party's real size as it is seated, whether booked or walking in, produces the most accurate count of the night. Changes at the door, such as a party of six that arrives as five, are corrected there and then.

Make it part of the host's routine, and check the count against the till at the end of the night. When the two agree, the venue's cover figures can be trusted.

Covers also shape the menu decisions. Knowing how many guests ordered each dish, rather than how many plates went out, shows what share of guests chose it. A dish ordered by a third of dinner guests is a signature; one ordered by one in fifty may be a candidate for change.

Our guide to menu mix covers how to read what guests actually order.

In the till, covers are usually entered when a table is opened: the server enters the number of guests on the check. That number is used to calculate spend per head and appears in the till's reports. It is only as accurate as the server's entry.

Common problems include checks opened with the default of one or two covers, tables that grow after the check was opened, and bar guests who never have covers entered at all.

Bar guests are one of the most common sources of disagreement. A guest who has a drink at the bar and leaves is a guest, but is it a cover? Some venues count them; many do not, counting only guests who sit down to eat in the dining room.

Neither is wrong, but the venue must decide and be consistent. A spend per head figure that sometimes includes bar guests and sometimes does not will move for no reason.

In the kitchen, the natural count is plates or dishes, not guests. A kitchen that sent out 350 mains last night knows it was busy, but it does not necessarily know how many guests ate. Some guests share; some order two starters instead of a main; children may share a dish.

When the kitchen talks about covers, it usually means the number of guests it was told to expect, from the bookings and the forecast. That is why the kitchen's covers can differ from both the booking system and the till.

Put the three together and the disagreements become clear. The booking system says 200 covers booked. The till says 176 covers on checks, because some walk-ins were never added and some tables were entered wrongly. The kitchen planned for 210, because the forecast added the usual walk-ins. All three numbers describe the same night.

None of them is a lie. They are different counts with the same name.

The fix is to agree, in writing, what the venue means by a cover, and where the official number comes from. A common definition is: a cover is a guest who was seated in the dining room, including children who ordered their own food, excluding staff meals, tastings and bar-only guests. The official count comes from the seated guests recorded in the bookings and walk-ins, or from the till, whichever the venue trusts more.

Once agreed, every report uses the same definition, and everyone knows which number is the real one.

Write the rules on one page: what counts as a cover, what does not, how walk-ins are recorded, how to enter covers on the till, how to handle a party that changes size, and where the official count comes from. Share it with the hosts, servers, managers and the accountant.

Train new staff on it. A new server who opens every check with one cover undoes the rules on the first night.

The covers report shows covers by day and service, against the same period last year, with the sources of guests, party sizes and no-shows. It is the starting point for understanding trade, and it is only useful if covers are counted consistently.

Our guide to the covers report, read on a Monday morning covers what it should show and the three questions it should answer.

Average spend per head is sales divided by covers: what each guest spent on average. It is one of the most useful numbers in a restaurant, and it depends entirely on the cover count. A wrong cover count gives a wrong spend per head.

Split it by service and by food and drink, rather than reading one figure for the week. Our guide to average spend per head, split properly explains why, and our guide to what is average spend covers how to raise it.

Covers per table is the average number of guests at each table: covers divided by tables seated. It shows how well the room's tables match its guests. A room of four-seat tables with an average of 2.3 covers per table is seating many couples at tables for four, which wastes seats.

Comparing covers per table with the mix of table sizes helps decide whether to change the layout. Our guide to designing a restaurant floor plan that sells covers how.

Booking capacity is the number of covers a venue can accept for a service, given its seats, its sitting length and its kitchen. It is not simply the number of seats. A 60-seat room with a two-hour sitting and a three-hour dinner service may be able to seat about 90 covers, if tables turn smoothly; a kitchen that can only cook for 70 at the peak limits it further.

Set booking capacity from the kitchen and the room together, and spread it across the service. Our guide to pacing covers so the kitchen never drowns covers how.

Table turns, the number of times a table is used in a service, connect covers to the room. A table for four that seats two couples in an evening has turned twice and served four covers. Our guide to table turnover rate covers how to raise turns without rushing guests.

Covers are also used to plan the floor. A server's section is often described in covers: twenty covers for an experienced server on a busy night, fewer for a new one, more in a simple café format. The number depends on the menu, the service style and the layout.

Use covers per server to build sections from the forecast, and watch the service on nights when a section was larger than usual. If guests waited too long, the section was too large.

Events and private dining count covers too, but they are planned differently. A private party of 30 is 30 covers booked as a single group, usually with a set menu and a confirmed number. The covers are known in advance, so the kitchen can prepare precisely.

Record event covers in the same way as ordinary covers, so they appear in the weekly totals. Many venues also report them separately, because their spend per head is often quite different.

In a group of venues, an agreed definition of a cover matters even more. If one venue counts bar guests and another does not, their spend per head cannot be compared, and the group's totals mean little. Agree the definition centrally, write it down, and make sure every venue follows it.

Our guide to multi-venue operations covers what a group should standardise.

Covers are most useful as a trend. Compare each week's covers with the same week last year, and each service with its own history. A gradual rise in covers on weekday lunches, or a gradual fall on Sunday evenings, tells the owner where the venue is winning and losing guests long before the accounts do.

Forecasts are usually built in covers, because covers drive the rota, the ordering and the prep. Start from covers booked for each service, add the usual walk-ins, and adjust for anything unusual. Our guide to a covers forecast your kitchen will believe covers the method.

Labour is often planned in covers per member of staff: how many guests each server can look after, or how many covers a cook can handle at the peak. These ratios differ by venue and format, but they turn a covers forecast into a rota.

The accountant may not use covers at all, but owners often add a covers line to the monthly figures, so they can see sales per cover, labour cost per cover and food cost per cover over time. These help explain the profit and loss in terms the floor understands.

A cover is the first term most new managers learn, but many others follow. Our glossary guides explain them in plain language:

Our restaurant glossary for new managers gathers sixty terms in one place.

Here is an illustrative Saturday and the arguments it caused. The figures are examples.

After the venue agreed one definition, seated guests including walk-ins, the official count for the night was 196 covers. Spend per head was calculated on 196, and the kitchen's forecast for the next Saturday started from it.

In servearound, a booking, a table and a check belong to the same guest, so covers are counted once. Reservation analytics shows covers day by day against the previous period, by meal period, by source and by party size, with no-show and cancellation rates, and the figures can be read with or without walk-ins. Bookings are followed from booked to confirmed, seated and completed, so covers booked and covers seated are both visible.

The dashboard shows covers and spend per cover against the previous period, and table analytics shows seatings, covers and utilisation for every table. See Analytics and the reservations module.

One guest who sits down to eat or drink. A table of four is four covers.

No. A table can hold several covers, and it may be used more than once in a service.

Most venues count children who order their own food. Decide your rule and keep to it.

Some venues count them, many do not. Decide and be consistent.

Because they count different things: covers booked, covers seated, walk-ins and covers entered on checks.

Covers booked are guests expected; covers seated are guests who actually arrived. The gap is no-shows and changes.

Sales divided by covers. A wrong cover count gives a wrong spend per head.

The number of covers a venue can accept for a service, given its seats, sitting length and kitchen.

A separate thing: a fee charged per guest in some venues and countries, for bread, the place setting or entertainment. It shares the word because both come from the place laid for each guest, but it is a price, not a count.

What is a cover? One guest, not one table. It is the unit behind the covers report, average spend per head, covers per table, booking capacity, forecasts and rotas. But the word is counted differently in three places: bookings, which count covers booked and seated; the till, which counts covers on checks; and the kitchen, which counts the guests it was told to expect.

Agree one written definition for the whole venue, record walk-ins, enter covers correctly on every check, decide whether bar guests count, and take the official number from one source. Once everyone counts covers the same way, the numbers stop starting arguments and start answering questions.