When operators tell me their catering channel is profitable, my first question is always the same. Show me the standalone P&L. About seven times out of ten, there is not one. There is a restaurant P&L with catering revenue and catering cost of goods folded into it, and there is a general sense that catering is helping. That is not a P&L. That is a hope.
I spent a stretch building an enterprise catering channel inside a $30M Michelin-recognized Bay Area group, delivering into Stanford, Google, Apple, Meta, LinkedIn, Salesforce, Cisco, Adobe, and Nvidia. Same kitchens as the restaurant service. Same executive chef. Same walk-in. Almost nothing else about the P&L behaved the same way. The operators who understand why get catering to punch above its weight. The ones who do not end up with a fast growing channel that quietly bleeds contribution.
Restaurant P&L is probabilistic. Catering P&L is deterministic.
Every other difference flows from this one. In a restaurant, you forecast covers based on historical data, weather, day part, and a general sense of the week. You staff a schedule against that forecast. You prep against a par level. You absorb the variance between your guess and reality with waste, with 86s, and with overtime.
In catering, every order arrives with a headcount, a menu, a delivery time, and a client, usually 48 to 72 hours before drop. Your food purchase, your prep sheet, your labor block, and your route sheet are all built to a number that is not a forecast. It is a contract. The variance around it is close to zero if the operation is disciplined.
Once you internalize that, every line on the P&L reads differently.
Fig. 1 · Percent of channel revenue on a healthy P&L, side by side.
Food cost: catering should run 4 to 6 points lower
A healthy full service restaurant food cost sits around 28 to 32 percent of revenue. A healthy corporate catering food cost should sit at 24 to 28. Same ingredients. Same menu. Lower cost of goods. Why?
- No portioning drift. Restaurant portions get eyeballed by a cook at ticket speed under pressure. Catering portions get weighed at prep against a spec, once, and packed.
- No 86 risk. The restaurant has to overbuy on a few items to protect a Friday rush against a stockout. Catering buys exactly to order.
- No plate return. Catering food that leaves the building either sells or gets logged as overproduction. Restaurant plates come back partially eaten and get scraped without a data trail.
- Cleaner walk-in count. Catering inventory turns are fast and deterministic. Restaurant inventory carries a longer tail of slow movers that quietly waste.
If your catering food cost is running at restaurant levels, something is broken. Usually it is a shared purchase order where the catering line item is guessed instead of specced, or a shared prep station where catering portions are done last with whatever is left.
Labor: the block versus the shift
This is the biggest single mistake I see. Operators run catering labor through the restaurant schedule. Meaning they staff a line cook for a full six hour shift because a catering drop lands at 11:45 a.m., and then they wonder why catering labor cost as a percent of catering revenue looks awful.
Catering labor is not a shift. It is a work block. A 200-person hot lunch drop needs, roughly, three people for a 90 minute prep and pack window, plus a driver for a 45 minute delivery and setup window. That is about 6.75 productive labor hours against, say, $3,600 of revenue. Labor as a percent of revenue on that job is about 5 percent, before you allocate any of the executive chef, expo, or dish time it consumed.
Even after you fully allocate the shared kitchen labor and add a fair share of overhead, a healthy corporate catering channel should run labor at 16 to 20 percent of catering revenue. That is 10 to 14 points below a healthy restaurant.
The catering P&L is a math problem. The restaurant P&L is a probability problem. Operators who treat them as the same business give up 8 to 12 points of contribution margin without noticing.
How to build catering labor properly
Catering labor sits in three buckets. Sales and account management. Kitchen production. Delivery and onsite setup. Each one has to be scheduled and costed differently.
- Sales and account management. A dedicated catering coordinator, salaried, whose cost gets allocated across the catering channel. In a channel doing $60K per month, one coordinator carries a labor load of roughly 6 to 8 percent of revenue and pays for themselves at half that if they are any good.
- Production. Catering prep gets built into the daily prep sheet the night before, using the same cooks the restaurant already has. The labor block is added to the day's schedule as a named window, not an assumption.
- Delivery and setup. A dedicated catering driver or a part time delivery team, paid hourly against the delivery window plus buffer. This is where operators overpay the most, usually by having a full time employee wait around for a two hour window.
The cost lines the restaurant P&L does not have
Catering brings its own cost lines that restaurant operators are not used to managing. Miss them and your reported margin is wrong.
Packaging
Chafers, sternos, disposable serving utensils, compostable plates, napkins, labels, delivery bags, and the cardboard boxes underneath all of it. In a corporate catering channel this typically runs 3 to 5 percent of catering revenue. It sits inside food cost on most systems, which is fine, but it has to be tracked. I have seen a channel report 26 percent food cost that was actually 31 percent once packaging got broken out.
Delivery cost
Vehicle, fuel, insurance allocation, driver labor, and any third party delivery fees. In a self delivered corporate channel this runs 2 to 3 percent of revenue. In a third party delivered channel this runs 8 to 12 percent, which is why third party delivery is almost never the right model above a certain volume.
Payment terms
This one is not a cost line, but it is a working capital line that behaves like one. Corporate catering clients pay net 30 or net 45. That means at any given time your fastest growing channel is also your largest accounts receivable, which is not free. On a $1M catering channel with 40 day payables, you are carrying about $110K of float. Charge for it or bake it into the price.
Contribution margin: the number that matters
Because catering avoids most of the fixed overhead that a restaurant has to carry, contribution margin, not net margin, is the correct number to run the channel on. Contribution margin equals catering revenue minus food, direct catering labor, packaging, and delivery.
A well-run enterprise catering channel produces 42 to 50 percent contribution margin on catering revenue. A healthy full service restaurant produces 28 to 34 percent. Same team, same building, and roughly 40 percent more contribution per dollar of revenue. That is the number that makes catering worth the operational complexity. That is also the number that gets destroyed the fastest when the channel is run out of the restaurant P&L.
How to price catering when the P&L is separate
Once the channel P&L stands on its own, pricing catering off retail menu prices stops making sense. A restaurant menu is priced against the restaurant's cost structure, its rent, its wage rates for a full service team, and the guest experience it provides. A catering order carries a completely different cost stack and delivers a different value.
The right way to price a catering item is to build up from its actual catering cost. Take the food cost of the item at catering portioning. Add packaging cost, which for a single serve entree with disposable plate, lid, utensil, napkin, and label runs 60 to 90 cents. Add the direct catering labor cost per unit at your operation's throughput. Add a delivery cost allocation based on the average drop size. Then apply a margin multiplier that targets the 42 to 50 percent contribution range.
For a typical hot lunch entree, that math lands the catering price 20 to 30 percent above the equivalent restaurant menu price for the same item. Which sounds like a lot until you look at the value: individually packaged, delivered onsite, set up on chafers, at a fixed time, for a client who is not comparing the price to what they would pay if they walked into your restaurant. They are comparing it to what a catering competitor would charge them, and to the productivity cost of their employees leaving the building for lunch.
Kitchen overhead allocation without a war
The one line item that consistently causes friction between the restaurant P&L and the catering P&L is shared kitchen overhead. Utilities, cleaning, small equipment repair, the executive chef's salary, dish room, walk-in space. Both channels use them. Neither channel wants to carry more than its share.
The fairest allocation I have used is one that both sides can agree to before the numbers land. Total the shared kitchen overhead for the month. Allocate it in proportion to gross revenue produced by each channel that month. It is not a perfect measure of actual usage, but it is a stable measure that both channel P&L owners can predict and neither can game. It also has the useful property that as catering grows relative to the restaurant, its share of overhead grows automatically, which is the right economic signal.
Split the channel P&L on day one
The single highest impact move an operator can make in a hybrid restaurant and catering business is separating the P&L cleanly, from the first month. Not at year end. Not when catering hits a magic revenue number. Day one.
That means a separate cost of goods bucket for catering food. A separate labor cost line for catering production, packing, and delivery. Packaging on its own line. Delivery on its own line. Sales coordinator allocated across the channel with a clear formula. A shared kitchen overhead allocation that both sides can agree is fair.
Once that split exists, the truth is visible. The restaurant P&L stops looking artificially bad because it is carrying catering food. The catering P&L stops looking artificially good because it is borrowing restaurant labor. And you can price catering off its actual economics, not off a retail menu that was built for a different business.
The point
Catering and restaurant sit inside the same building for a reason. They share fixed costs, they share equipment, they share people, and the operator who can run both from one kitchen has a real economic advantage over the operator who cannot. That advantage only shows up if you treat the two P&Ls as the two different games they are.
Deterministic on one side. Probabilistic on the other. Separate the P&L, split the labor into blocks not shifts, put packaging and delivery in their own lines, and manage the channel on contribution margin, not net. Do that and catering becomes what it should be, which is the highest margin dollar the same kitchen can produce.