Pacing covers so the kitchen never drowns

Pacing covers is the craft of deciding how many guests sit down, and order, in each part of the evening. Every kitchen has a ceiling: a number of tickets it can cook well in a given quarter of an hour. Below that ceiling, food goes out hot and on time. Above it, tickets stack up, the pass fills with plates waiting for their partners, and the dining room starts to feel the delay one table at a time. Pacing keeps arrivals below the ceiling, so the night stays good from the first table to the last.

Most kitchens that drown do not drown because the room is too big or the night too busy. They drown because too many of the night's covers arrived at the same moment. A room that seats 80 can serve 160 guests on a Saturday comfortably if they arrive in a steady stream, and struggle with 60 if 40 of them sit down between eight and a quarter past.

This guide explains what pacing covers means, why bookings cluster, how booking capacity differs from kitchen capacity, how to find your own ceiling from your own tickets, and how to turn it into booking slots and table pacing that hold up on the busiest night of the week. It also covers what to do on the night itself, when the plan meets the guests who actually turn up.

Pacing covers means controlling how many guests are seated in each interval of a service, usually each fifteen minutes. A pacing limit might say: no more than 20 covers seated between 7:00 and 7:15, and no more than 20 between 7:15 and 7:30. It is a limit on arrivals, not on the total for the night. A room can take as many guests as its tables and sittings allow, provided they do not all arrive together.

The reason is simple. Guests who sit down at the same time tend to order at the same time, and their tickets reach the kitchen together. Ten tables seated at 8:00 produce ten drinks orders at 8:05, ten food orders at 8:15 and, if nothing intervenes, ten sets of mains fired at around 8:45. A kitchen that cooks comfortably for five tables at once is now cooking for ten, and every one of those tables waits.

Pacing is not the same as turning tables faster, and it does not mean fewer guests. Done well, it lets the room serve more guests, because a kitchen working at a steady rate all evening cooks more plates than one that sits idle at six and drowns at eight. Our guide to table turnover rate covers the other half of the evening: how long each table stays once it is seated.

Left alone, bookings bunch. Guests want the popular times, usually somewhere between half past seven and half past eight on a weekend, and a booking page that offers every slot to everyone will fill those slots first. Phone bookings add to the cluster, because the person answering often accepts the time the guest asks for rather than suggesting the quarter hour before or after.

Party size makes the cluster worse. A table for eight at 8:00 is not one ticket; it is eight starters and eight mains arriving at once, often with a wider range of dishes than a smaller table orders. Two large parties booked within a few minutes of each other can take a station from comfortable to overwhelmed on their own.

Walk-ins and late arrivals add noise on top. A party booked for 7:30 who arrive at 7:55 are now part of the eight o'clock rush, through no fault of the diary. Pacing has to allow for this. It cannot assume every guest arrives on the minute, and a plan with no slack in it will fail on the first busy night it meets.

Booking capacity is the number of guests the room can seat at any one time: the tables, their seats and the ways they combine. Kitchen capacity is the number of covers the kitchen can cook well in a given interval. They are different numbers, measured in different ways, and confusing them is the most common reason a full room produces a bad night.

A room's booking capacity is usually far higher than the kitchen's capacity per interval. A 70-seat room could, in theory, seat every one of its guests at 8:00. No kitchen built for that room could cook 70 covers in the next fifteen minutes. A booking diary left to itself only knows about tables, so it will happily fill the room to its booking capacity at the worst possible moment.

That is why pacing has to be set deliberately. The tables tell you how many guests can sit down; the kitchen tells you how quickly they can be fed. Booking slots and table pacing are where the two numbers are reconciled, before the guests arrive rather than in the middle of service.

The best source for your kitchen capacity is your own tickets. Take a few busy services and count how many covers' worth of food the kitchen sent in each fifteen minutes, then look at ticket times in the same intervals. You are looking for the point where ticket times start to climb: the number of covers per interval at which the kitchen stops keeping up.

In most kitchens there is a clear bend. Up to a certain number of covers per quarter hour, ticket times hold steady. Beyond it, they rise quickly, because each new ticket waits behind the ones in front of it. The behaviour is the same one described by queueing theory: when work arrives faster than it can be done, the queue grows, and waiting times grow with it. That bend is your ceiling.

It is usually lower than the chef's instinct suggests, because instinct remembers the nights that went well and forgets the ones that did not. Measure it by meal period and by menu. A lunch menu of simple dishes may have a much higher ceiling than a dinner menu with complex mains, and a tasting menu has a ceiling of its own. When the menu changes with the season, measure again.

A kitchen's capacity is set by its busiest station, not by its average. If the grill cooks most of the mains, the grill sets the ceiling for the whole kitchen: the fryer and the cold section may have room to spare, but the tables are still waiting for their steaks. Find the station that falls behind first, and you have found the station that sets your pacing.

Ticket times by station show this clearly. If one station's prep times rise sharply on busy nights while the others hold steady, that station is your bottleneck. You can pace around it, by limiting the covers seated in each interval, or you can widen it: an extra cook on the grill on Saturdays, a dish moved to another station, or more preparation done before service.

The pass is a bottleneck too. Even when every station keeps up, a single person calling and checking every plate can only send so many tables in a quarter of an hour. If plates are dying under the lamps while the pass catches up, the pass is the ceiling, and pacing has to respect it as much as any station.

Once you know your ceiling, turn it into booking slots. The simplest approach is to decide how many covers you will seat in each fifteen-minute interval and offer only the times that keep you below that number. If your kitchen copes comfortably with around 18 covers per quarter hour, the diary should not offer 7:45 to a party of six once 14 covers are already booked into 7:45.

Staggering is the other tool. Instead of offering every table at every time, offer some tables on the hour and some on the half hour, or start the larger tables slightly earlier than the smaller ones. Spreading the first seatings across the first ninety minutes of service gives the kitchen a steady flow from the start rather than a wave at the end.

Keep the popular times precious. The 8:00 slot on a Saturday will always be in demand, so let it fill with the parties that suit it, such as couples and small tables, and steer larger groups to earlier or later times when they book. Guests are usually happy to move fifteen minutes if you ask before they have their heart set on a time.

Table pacing is the same idea applied on the floor: seating tables in a steady sequence across sections, so no server and no part of the kitchen receives a burst. Even with bookings spread across the evening, a host who seats four arriving parties in one section at once has created a small rush of their own making.

Seat across sections in turn, give each server time to greet one table before the next arrives, and hold a waiting party at the bar for a few minutes if a section has just been seated. A drink at the bar at the start of the evening is a better experience than a table where nobody arrives for ten minutes because the server is greeting three others.

Table pacing also applies to the order in which tables are fired. If the server holds the mains and fires them when starters are nearly finished, the kitchen receives each table's mains when they are needed, not all at once when the tables first order. That spreads the load across the evening and keeps mains from arriving while starters are still being eaten.

Large parties deserve their own pacing rules, because one of them can use a station's capacity for a whole quarter hour. Many venues set a separate limit for large tables, such as one party of eight or more in any half hour, and keep them away from the busiest intervals altogether.

Ask large parties to choose dishes in advance where the format allows it, or offer a set menu or a short choice so the kitchen can plan. Seat them a little earlier than the main rush, take their order quickly and fire each course so the whole table eats together. A large party ordered and fired before the eight o'clock rush is a large party that will not swamp it.

Treat large parties as a pacing question when they book, not when they arrive. A booking page or a phone script that steers a party of ten to 7:00 instead of 8:00 has done more for the kitchen than anything the host can do on the night. Our guide to designing a restaurant floor plan that sells covers how to plan the table combinations large parties need.

Pacing does not end with the booking diary. Walk-ins arrive whenever they like, and a run of them at eight o'clock can undo a carefully paced evening. Treat walk-ins as part of the same plan: seat them where the next interval has room, and offer a short wait at the bar where it does not.

A live waitlist helps, because it lets the host see who is waiting and choose the moment to seat them rather than seating everyone the instant a table frees. Quoting an honest wait, and seating the party when the kitchen can take them, is better than seating them at once and leaving them waiting for food. Our guide to waitlist management covers how to quote a wait and keep it.

No-shows and late cancellations change the pace as well. A table that does not arrive at 7:30 leaves a gap in the kitchen's load that can be filled from the waitlist. Our guide to what a no-show costs you covers that cost, and how a waitlist recovers part of it.

Pacing fails when bookings arrive through channels that do not know about each other. If the phone, the booking page and the host's notebook each take bookings into a different list, nobody can see how many covers are booked into 8:00. The only way to pace reliably is to keep every booking in one book, with the same view for everyone who takes them.

That book also needs to know about the tables. A booking slot is not just a time; it is a time at a table that will be free. A diary that knows which tables are joined for a large party, and which are held for the host to assign, will only offer the slots the room can actually deliver.

It also needs to know about the guests who are booked but not yet confirmed. Reminders, and a simple way for guests to change or cancel, turn many would-be no-shows into cancellations you can resell. Our guide to booking reminders covers how many to send and when.

Here is an illustrative Saturday for a 70-seat room with a ceiling of around 18 covers per fifteen minutes, found from its own ticket times. The figures show the method, not a target for your venue.

The point is not the specific numbers, which depend on your kitchen, your menu and your team. It is the method: measure the ceiling, offer only the slots that respect it, keep large parties out of the rush and fire courses when tables are ready for them.

Even the best-paced diary meets reality at the door. Guests arrive early, late and in different numbers from the booking. The host's job is to keep pacing covers on the night: watching how many guests have been seated in the last fifteen minutes, how the kitchen is coping, and when to hold the next party at the bar for a few minutes.

Keep a short line of communication between the pass and the host stand. A chef who can say "give me ten minutes" before the kitchen falls behind lets the host slow the seating before tickets pile up, not after. A host who can say "two large tables just sat" lets the kitchen get ready for them.

If you take bookings that need confirming, a busy Saturday is the night to confirm them deliberately. Looking at each new booking against the covers already booked into its interval lets you call the guest and suggest a different time before the problem ever reaches the kitchen.

Pacing protects the kitchen from its ceiling; raising the ceiling makes the protection less necessary. The biggest single lever is preparation. A kitchen that starts service with everything prepared, portioned and to hand cooks faster under pressure, which is the whole point of mise en place. Our guide to the prep list covers how to plan it.

The menu is the second lever. Dishes that take a long time, or that all land on one station, lower the ceiling. A busy-night menu that spreads the work across stations, with fewer dishes that need last-minute attention, can raise the number of covers the kitchen cooks per interval without adding a cook.

The third lever is people. An extra cook on the bottleneck station, or a dedicated person on the pass, widens the narrowest point in the kitchen. Measure before and after, so you know whether the extra cost raised the ceiling enough to pay for itself.

Pacing can be measured. Look at ticket times through the evening, the share of tickets that reached the pass on time, and how many dishes were remade. If pacing is working, ticket times stay level across the evening rather than climbing after eight. If they still climb, either the ceiling is set too high or bookings are slipping past it.

Compare prep times by station to see which part of the kitchen struggles first, and read the median and the slowest tickets separately. The median shows a typical ticket; the slowest tenth shows the tables that had a bad night. A pacing plan that improves the median but leaves the slowest tickets unchanged has not solved the problem.

Ask the floor as well. Servers know which tables waited and why, and a short conversation after service often explains what the numbers show. Pair it with a look at covers by hour across the week, so you can see which services are reaching the ceiling and which have room to take more.

In servearound, booking rules set the service periods and the times offered inside them, how long a sitting lasts, the largest party and the notice you need. Each table takes online bookings or not, for the party sizes and on the days and hours you choose, so you can stagger which tables are offered when, and keep a table for the host to assign by hand. When a time is full, the booking page offers the waitlist instead.

Online bookings can be confirmed the moment they are made or arrive pending until somebody confirms them, and a table taken on the phone is no longer offered online, so every booking lands in one book. See the reservations module.

In the kitchen, dishes sit on the check until they are sent, so servers can send the starters, hold the mains and fire them when the table is ready. Tickets go to the station that cooks them, oldest first, each counting its own time in the kitchen. Analytics shows prep times by station and covers by day and hour, so you can find your bottleneck and your busiest hours from your own services. See fulfillment for how the floor and the kitchen share one order.

There is no general answer. It depends on your kitchen, your menu and your team. Measure it from your own ticket times on busy services, and set the limit just below the point where ticket times start to climb.

Usually not. It means spreading the same bookings across the evening. On the busiest nights it can mean declining 8:00 and offering 7:30 instead.

Offer the nearest times that have room, and the waitlist for the time they wanted. Many guests will move fifteen minutes; those who will not can wait for a cancellation.

Yes, if lunch is busy. Lunch guests often have limited time, so a kitchen that falls behind at one o'clock costs more goodwill at lunch than it would at dinner.

By staggering start times so the same course is not being plated for every table at once. Tasting menus are often easier to pace, because every table orders the same dishes.

They use the same capacity. Count them in the same intervals as the dining room when you measure your ceiling, or they will quietly push the kitchen over it.

Whenever the menu, the team or the kitchen changes, and at least every few months. A new dish on the grill can change it overnight.

Pacing covers means limiting how many guests sit down in each interval of the evening, so the kitchen never receives more tickets than it can cook well. Booking capacity is what the room can seat; kitchen capacity is what the kitchen can cook per interval, and its busiest station sets it. Find your own ceiling from your own ticket times, then turn it into booking slots that respect it, stagger the tables and keep large parties out of the rush.

On the night, pace the tables across sections, fire courses when tables are ready, seat walk-ins when the kitchen has room, and keep the pass and the host stand talking. Measure ticket times through the evening, and raise the ceiling with preparation, a better spread of dishes across stations, and people at the narrowest point.