Running a central kitchen for three venues

A central kitchen is one production kitchen that cooks for several venues. Stocks, sauces, doughs, braises and portioned proteins are made there in bulk, chilled, and sent out to the sites, which finish and plate them during service. For a group with three venues, it can be the difference between three kitchens each doing the same morning's prep and one team doing it once, properly, while the site kitchens concentrate on service.

It can also be an expensive mistake. A central kitchen adds rent, equipment, a vehicle, a second set of food safety procedures and a layer of ordering and paperwork that did not exist before. It changes what the chefs at each venue do all day, and not every chef thanks you for it.

This guide covers when a central kitchen starts to pay, what to make there and what to leave to the sites, and how to run the four things that decide whether it works: the kitchen production plan, internal ordering, the transfers between kitchens, and central kitchen costing. Every figure in it is an example.

The terms vary. Some groups say central production unit, some say commissary kitchen, some say hub kitchen. The idea is the same: food is prepared in one place for consumption in several others, usually under cook-chill or cook-freeze, and the sites are supplied on a schedule.

It helps to separate three set-ups that are often lumped together:

Most groups with three venues start with the second option and move to the first when the flagship kitchen starts to buckle. That is a reasonable path, as long as everybody knows that the flagship's kitchen now has two jobs and is resourced for both.

The case for a central kitchen rests on a simple observation: three kitchens making the same sauces, stocks and bases every morning are paying three times for work that could be done once. Batch production uses labour more efficiently, makes better use of equipment that would otherwise sit idle, and produces a more consistent result.

The costs are just as concrete. A worked example for a group of three venues, with invented figures, shows how the sums come out.

Suppose each venue's kitchen spends about 25 hours a week on preparation that could move to a central kitchen: stocks, sauces, dough, braising, butchery and portioning. That is 75 hours a week across the group. In the central kitchen, working in batches with better equipment, the same output might take 45 hours, a saving of 30 hours a week. At an example cost of 18 an hour including on-costs, that is 540 a week, or about 28,000 a year.

Against that, the central kitchen might cost, in the same example, 20,000 a year in rent and running costs for a small unit, 6,000 a year for a van and its running costs, and a part-time driver at 9,000 a year. That is 35,000, more than the labour saving alone.

So in this example, labour savings do not pay for a central kitchen on their own. What tips the balance, when it tips, is usually some combination of the following:

The honest test is to put numbers on each of those for your own group, with invented optimism taken out, before signing a lease.

A central kitchen tends to make sense when several of these are true:

It tends not to make sense when:

If you are still at the stage of opening a second venue, our guide to what breaks first with a second venue covers the problems that usually come before a central kitchen is worth considering. The wider question of what to centralise, from buying to marketing, is in our guide to multi-venue operations.

The decision for each item comes down to four questions. Does it take a lot of labour? Is the recipe stable? Does it travel and hold well chilled? Does making it in larger batches improve it, or at least not harm it? Items that answer yes to all four belong in the central kitchen.

Good candidates:

Items better left to the site:

Between the two sits a grey area: items that could be made either way. Start by keeping them on site, and move them only when the central kitchen has proved it can deliver the core items reliably. A central kitchen that tries to make everything in its first month usually makes nothing well.

The central kitchen does not cook to orders as they arrive. It cooks to a plan, and the plan starts with what the venues expect to sell.

Build the kitchen production plan in four steps:

The plan should run several days ahead for long-cook and long-life items, and a day ahead for items with a short shelf life. Review it daily against what the venues actually used, because a forecast that is never checked becomes a habit, not a plan.

At each venue, the same thinking runs one level down, in the daily prep list. Our guide to the prep list that ends the morning guesswork covers how a site kitchen plans the finishing work once the central kitchen's deliveries arrive.

Internal ordering is how each venue tells the central kitchen what it needs. Without a clear system, the central kitchen hears requests by phone, text and shouted message, and production becomes reactive.

A workable system has a few firm rules:

Track two figures from the start: how often orders arrive after the cut-off, and how often an emergency top-up is needed. Both tell you whether the forecasting and the pars are working.

Every delivery from the central kitchen to a venue is a transfer. It moves food, and it moves value: stock leaves one kitchen's books and arrives in another's. Both sides of that need recording.

On the food side, each delivery needs a delivery note listing what was sent, in what quantity, with batch numbers or production dates, and use-by dates. The venue checks the delivery against the note on arrival, records the temperature of chilled items, and signs. Any shortfall or rejected item is noted there and then.

On the value side, each item has a transfer price, and the delivery note's total is what the venue is charged. Without transfer prices, the central kitchen's costs sit in one place and the venues' food cost percentages look artificially low, or the costs are spread by guesswork at the end of the month. With them, each venue's food cost reflects what it actually used.

Count transfers in stock counts too. Food that has left the central kitchen but has not been recorded as arriving at a venue is the commonest reason for a group's combined stock figures not adding up. Our guide to running a stock count properly covers the counting itself.

Central kitchen costing has two jobs. It tells each venue what its food really costs, and it tells the group whether the central kitchen is earning its keep.

Start with the recipe cost of each batch: every ingredient at its purchase price, divided by the yield of the batch, gives a cost per kilogram or per portion. Our guide to recipe costing walks through yields and trim losses.

Then decide what the transfer price includes. There are three common choices:

An example, with invented figures. A batch of braise uses ingredients costing 96 and yields 12 kilograms, so the ingredient cost is 8 a kilogram. If the central kitchen's labour and running costs come to an example 1,200 a week, and it produces 600 kilograms a week across all items, those costs add 2 a kilogram. The transfer price is 8 at ingredient cost or 10 at full cost.

Whichever method you choose, use it consistently, and make sure everybody reading a venue's food cost percentage knows which one it is. Our guide to calculating food cost percentage covers the rest of that calculation.

A central kitchen extends the food safety chain. Food is cooked in one place, cooled, stored, transported, stored again and reheated or finished in another. Each of those steps is a point where things can go wrong, and each needs a control.

Your food safety management system, based on HACCP principles, should cover at least:

In the UK, the Food Standards Agency's guidance on food safety management systems explains what such a system needs to show. The general principles of cook-chill production are summarised in Wikipedia's article on cook-chill, but your own procedures must be built for your own food and premises.

A central kitchen is a food business in its own right and must be registered with the local authority like any other. Depending on what it makes and who it supplies, it may also need approval before it starts.

In England, Wales and Northern Ireland, some establishments that handle products of animal origin, such as meat products and some meat preparations, and supply them to other businesses, need approval from their local authority rather than just registration. Whether that applies to a kitchen supplying your own venues depends on what it makes, in what quantities and how it supplies them. The Food Standards Agency's page on applying for approval of a food establishment lists the establishments concerned. Talk to your local authority's food team early, before the kitchen is fitted out, because approval can affect the layout. Elsewhere, check the rules where you trade.

A central kitchen is only as good as its deliveries. Plan them as carefully as the production.

Start with one delivery a day. A second daily run is a solution for a specific problem, such as short-life items or a lunch rush, not a starting point.

A central kitchen changes jobs. The site chefs no longer make their own stocks and sauces, and some of them will feel that the craft has been taken from them. The central kitchen team cooks food it never sees served. Both groups need to understand why the change is being made and what it gives them.

A few practices help:

Pick a small number of figures and review them every week, from the first week:

The last two matter most. The point of the central kitchen is lower total cost, or more capacity at the same cost. If combined labour and food cost have not moved after a few months, find out why before expanding it.

A central kitchen is mostly a production and logistics problem, and the production plan, the transfers and the costings stay in your own system today. servearound helps with the parts that start at the venues.

Each venue in a group sits under one business in servearound, with a network dashboard across all of them. The menu report in Analytics shows units sold for every dish at each venue over any range of dates, which is the history behind a dish forecast and therefore behind the kitchen production plan. At each venue, orders from the point of sale go to the kitchen display by station and course, so the finishing work on site runs from the same tickets every time.

servearound does not plan production, take internal orders between kitchens, record transfers or hold recipe costs today. Inventory tracking, for the stock behind the menu, is coming. To see how a group of venues runs on servearound, look at pricing or get in touch.

There is no fixed number. Some groups make it work with two venues close together and a shared menu; others with five venues and different concepts never do. Work out the labour saved, the waste and buying gains, and the costs of premises, transport and people for your own group before deciding.

In everyday use, nothing much. Both describe a kitchen that prepares food for consumption elsewhere. Some people use commissary kitchen for shared or rented production space, and central kitchen for a group's own.

The items that are labour-heavy, stable, travel well and are used at every venue: stocks, sauces, doughs and braises are the usual start. Add more only when those arrive reliably.

Choose ingredient cost only, ingredient cost plus a share of the central kitchen's costs, or a set price reviewed periodically, and apply it consistently. The important thing is that each venue's food cost reflects what it used and that everybody knows which method is in use.

Yes. It is a food business in its own right, and its system must cover cooling, labelling, transport and allergen information as well as cooking. It must also be registered, and depending on what it makes and supplies, it may need approval.

Several days for long-cook and long-life items, a day ahead for anything with a short shelf life. Review it daily against what the venues actually used.

It can, through less waste and better buying, but only if transfers are priced and recorded. Without transfer prices, the venues' figures may look better while the group's total cost has not changed at all.

Underestimating logistics and ordering. The cooking is rarely the problem. Late orders, missed deliveries and food that arrives without labels or allergen information are what make site teams lose faith in it.

A central kitchen makes stocks, sauces, doughs, braises and portioned items for several venues, which finish and plate them on site. It pays when labour savings, less waste, better buying, consistency and room to grow outweigh the premises, transport and extra food safety work, and labour savings alone rarely do it. Work out those sums for your own group, with the optimism taken out.

If you go ahead, make centrally only what is labour-heavy, stable and travels well, and leave texture-sensitive cooking to the sites. Run a kitchen production plan from each venue's forecast and pars, take internal orders one way with a firm cut-off, record every transfer with a transfer price, and keep the food safety chain intact from cooling to reheating, with allergens recorded all the way. Measure accuracy, waste and combined cost from the first week.