How to price a catering event so you actually make money

There is a version of catering pricing that everybody learns first and almost nobody survives on: work out the food cost, multiply by three, quote that. It feels like a business. It is actually a way of losing money slowly enough that you do not notice for two years.

The problem is not the multiplier. The problem is that the multiplier is standing in for three costs nobody wrote down, and those three costs move independently of the food.

The four costs in every event

1. 1Raw food cost. What the ingredients cost you, at the yield you actually get, including the trim you throw away.
2. 2Direct labour. Every hour that exists because this event exists: prep, load-in, travel, service, breakdown, and the drive home. Plus payroll burden.
3. 3Event-specific costs. Rentals, disposables, linen, fuel, mileage, parking, permits, event insurance, and the credit card fee on the deposit.
4. 4Your share of overhead. Rent, insurance, software, your own salary, the van payment, the accountant. This event's slice of the cost of existing.

Only the first one has a multiplier folklore attached to it. That is why the folklore fails: it prices one cost and hopes the other three come out in the wash.

Cost 1: food, at real yield

The number that matters is not what the invoice says, it is what lands on the plate. A case of whole chickens at $2.10/lb is not $2.10/lb of chicken meat — after bone and trim you are closer to $4.60/lb of usable protein, and if you cost the recipe at $2.10 you have just built a 50% error into every quote that uses it.

Yield percentages are the least glamorous number in this trade and the one that quietly decides whether you make money. Write them down once per ingredient, use them forever, and update the prices — not the yields — when the invoice changes.

Cost 2: labour, including the hours you keep forgetting

Ask a caterer how many labour hours went into a 100-guest dinner and you will usually get the service hours. Here is the actual list.

Twenty-four of those sixty-three hours are the ones people picture. If you priced service hours only, you priced 38% of your labour.

Then add payroll burden. Employer taxes, workers' comp, any benefits, and the paid time that is not billable. Depending on where you operate that is 15-25% on top of the wage. A server you pay $22/hour costs you around $26-27/hour before you have made a cent.

Which is why the rule of thumb for billing labour is 2.5-3x the wage. Not because you are marking up people, but because the third of that number that looks like profit is burden, unbilled hours, and overhead.

Cost 3: everything else the event drags in

• Rentals — and the delivery, pickup and damage waiver, which is where rental quotes get away from you
• Disposables, foil, cling film, sterno, ice
• Linen and its laundering
• Mileage and fuel at a real per-mile rate, not just the gas
• Parking, loading permits, venue fees, union labour at some venues
• Event insurance or the certificate the venue demands
• Card processing on the deposit and the balance, usually 2.9% + 30c
• Tastings — a tasting for four is a small catered event you did for free

That last one deserves a policy. Free tastings for every enquiry is a lovely way to work sixty hours a week for people who were never going to book. Charge for it and credit it against the booking; the people who are serious will not blink, and the people who blink were the problem.

Cost 4: overhead, allocated honestly

Total your annual fixed costs — rent, utilities, insurance, software, vehicle, your salary, professional fees, marketing. Divide by the number of events you realistically do in a year. That is what each event has to carry before it contributes anything.

Do it once a year and it will be uncomfortable. $96,000 of overhead across 80 events is $1,200 per event that has to come from somewhere. If you have been quoting on food-times-three, that $1,200 has been coming out of your own pay.

The worked example

One hundred guests, plated dinner, mid-range menu, venue twenty minutes away.

Now price it. At a 20% net margin you divide by 0.80, not multiply by 1.20 — a mistake that costs you money every single time:

• $5,574 ÷ 0.80 = $6,968, which is $69.68 per guest.
• Food cost as a share of revenue: $940 ÷ $6,968 = 13.5%.

Look at that food cost percentage. If you had priced this on food-times-three you would have quoted $2,820, about $28 a head, and lost roughly $2,750 on the night while feeling busy and successful. That is the whole trap in one line.

Margin targets that hold up

Weddings carry a higher margin for a real reason, not a cynical one: the coordination load, the change requests, the rehearsal, and the fact that there is no second attempt. You are pricing risk and attention, and both are genuinely more expensive.

Three pricing decisions worth making once

Put a minimum on it

A minimum event value is the fastest profitability fix available to most caterers, because small events consume nearly the same fixed hours as large ones. Twenty-guest events at $40/head do not scale; they just fill your calendar with breakeven.

Charge for the change

Final guest count due 10 days out, billed at the greater of confirmed or actual, and menu changes after that carry a fee. Not to punish anybody — to stop the fourth revision being free labour.

Show fewer numbers, not more

Quote a per-guest price and a small number of clearly-named line items. An itemised sheet showing $4 for bread invites a negotiation about bread. Your costing can be forensic; your proposal should be confident.

The part that has to be automatic

All of the above is arithmetic you can do in a spreadsheet, and the first three times you will. Then chicken goes up 11%, you have forty live menus, and the spreadsheet becomes a museum piece — right on the day it was built and quietly wrong ever since.

The version that survives is one where ingredient prices live in one place, recipes roll up to a real plate cost, labour is estimated from the actual schedule rather than a guess, and the proposal takes its numbers from those. That is what we built Apron Desk to do, and it is the only part of this article that requires software rather than discipline.