Chapter 29: Catering, Private Events, and Banquets: Selling the Room and the Kitchen by the Head

"You do not sell a party. You sell the promise that nothing will go wrong in front of forty people the client had to invite." — constructed; the sentence every event salesperson eventually learns to say out loud

Overview

Here is the operating question, and it is not the one most operators think it is.

You have a 68-seat dining room. On Saturday at 7:30 it is full, the hearth is running over rate, and you are turning people away. On Tuesday at 7:30 there are thirty people in it and four of your seventeen tables have never been sat. Chapter 24 measured that asymmetry precisely: the room runs 43% empty on Saturday across the whole service window while the fire runs over rate, and Tuesday and Wednesday are demand-constrained — not capacity-constrained, not kitchen-constrained. Demand-constrained. Nobody is coming. There is no pricing trick, no menu change, and no promotion that reliably conjures forty walk-ins on a February Tuesday in a neighborhood where the offices emptied at five.

Events are how you sell the hours the dining room cannot.

That is the whole argument, and it is worth stating flatly because the industry usually sells events on a different and weaker claim — that they are "high margin." Some of them are. Many of them are not. A Saturday buyout at what feels like an enormous number can lose you money against simply unlocking the door, and this chapter will show you the arithmetic that proves it, to the dollar. What events genuinely offer is something else and something better: known covers, a known menu, and money collected in advance. They remove almost every source of uncertainty a restaurant normally lives with. That is a real, quantifiable advantage, and it is not the same thing as a higher margin.

The Bellwether plan carries a line item you now own: fourteen private events at an average of \$3,000 each, \$42,000 of revenue — the largest component of the revenue bridge after the patio, on a \$1,550,000 year. By the end of this chapter you will have costed one of those events to the cent, priced the rest against the nights they occupy, and either defended that \$42,000 or replaced it with a number you can put your name on.

You will also learn to write the one document that separates a profitable event business from a stressful one: the banquet event order. It is a page and a half. It prevents essentially every argument that events generate. Most independents do not write one, and most independents have a story about the wedding that ended badly.

In this chapter, you will learn to:

• Say exactly what is different about event economics — and name the two things that are not different, which is where operators lose money.
• Decide whether a restaurant without a private dining room should sell buyouts at all, and at what price, using displacement rather than instinct.
• Build a published price sheet: per-head packages, food-and-beverage minimums by night, and site fees.
• Write a banquet event order that a chef, a bartender, and a captain can all execute from without asking a question.
• Structure a deposit and cancellation ladder that protects capacity you have already sold.
• Handle service charge versus gratuity in a contract correctly — including what it costs you in payroll tax and overtime.
• Cost an event to a true contribution margin, net of the à la carte covers it displaced.
• Decide whether to take your kitchen off-site, and what you are accepting when you do.

Learning Paths

🏗️ Opening — §29.2 and §29.3 are yours: what your specific room can sell, and what you should publish for it. Do not put an event line in a pro forma until you have done §29.7 for at least one modeled event. 📋 Managing — §29.4, §29.5, and §29.9. The BEO and the deposit ladder are Monday-morning documents. §29.9's response-time discipline is the single cheapest revenue lever in the chapter. 🍸 Beverage — §29.3's package-versus-consumption decision is the largest single variable in event profitability, and §29.6 determines who owns the money your bartender earns. §29.8's off-premise alcohol rules are the ones people actually get in trouble over. 🚚 Small Format — most of a truck's real money is private gigs (Chapter 30), and every pricing, deposit, guarantee, and BEO structure here transfers intact. §29.8 is your core operating chapter, not a footnote.

29.1 Why event margin is different: known covers, known menu, prepaid

Start with what a restaurant normally does not know at 4:00 p.m.

It does not know how many people are coming. Bellwether's Tuesday forecast is 62 covers, and the honest error band on a mid-week forecast in year one is on the order of ±15% — call it 53 to 71 people. It does not know what they will order, so it holds inventory for a distribution rather than for a list. It does not know when they will arrive, so it staffs for a curve. It does not know whether it will be paid at the end of a table (it almost always will, but the money arrives after the food is bought, cooked, and eaten). And every one of those unknowns is expensive: the buffer inventory, the extra half-shift, the prep that gets thrown away Thursday, the cook standing at a cold station at 8:45.

An event removes all four.

Covers are known. A signed guarantee is a contractual number, not a forecast. You are told, in writing, at a stated deadline, how many people will eat. If fewer show up, you bill on the guarantee anyway. Forecast error goes to zero — not "small," zero — for that portion of the night.

The menu is known. Not "a menu" — the menu, with counts. Twenty-four hearth chickens and sixteen charred cabbages, confirmed eight days out. You order twenty-four chickens. Compare that with à la carte, where the same forty guests would have produced an unknowable menu mix and you would have carried par for every item on the board.

The money arrives early. Deposits and progress payments land before the food is purchased. This is the inverse of the working-capital problem Chapter 33 is built around. An à la carte cover is financed by you; an event cover finances you.

Waste is near zero. Nothing is 86'd, nothing sits, nothing gets thrown out on Thursday because Tuesday was slow. Bellwether's event food cost, worked in §29.7, comes in at 27.0% against a 30% à la carte food-cost target — and that three-point gap is not clever purchasing. It is the absence of uncertainty, converted directly into cost.

Labor is schedulable. You know the count nine days out, which means the schedule is written to the work instead of to a guess. Chapter 19's staffing guide is a probabilistic instrument; an event schedule is a deterministic one.

🧮 Run the Numbers

What "prepaid" is actually worth.

Fourteen events at \$3,000 of food and beverage each is \$42,000. Add a 22% service charge and an illustrative 7% sales tax and the total the clients actually pay is:

\$42,000 × 1.22 = \$51,240, × 1.08 = \$55,339.20 of gross billing across the year.

Under the deposit ladder in §29.5, roughly half of each invoice is collected before the event — about \$27,670 — and it arrives on average some forty-five days ahead of the cost it pays for.

That is not profit. Not one cent of it is profit. It is timing, and in a business where the difference between open and closed is frequently a Thursday in February when payroll clears (Chapter 1's fourth killer, Chapter 33's whole subject), timing is worth something real. An event book with money in it is a cushion that does not appear anywhere on a profit-and-loss statement.

The trap inside this: deposit money in the operating account looks like money you have. It isn't. It is money you owe as food and labor on a date already in the calendar. Operators who spend deposits are the operators who cannot buy the food in October for the party they sold in June. Hold deposits as what they are — a liability with a due date.

The two things that are not different

This is where the chapter earns its keep, because both of the following are said constantly and both are wrong.

False claim one: "events are higher margin." They are not automatically higher margin. Run the canonical Bellwether event all the way down in §29.7 and it produces a 40.4% contribution on food and beverage revenue, before occupancy and general overhead. Run \$3,000 of ordinary à la carte through the plan's own cost structure — 27.8% blended cost of goods, 32.3% all-in labor, roughly 3% card processing — and it produces about 36.8%. So the event is modestly better, and only because it carries none of the salaried kitchen and management labor the à la carte number absorbs. Strip the comparison down to genuinely variable cost on both sides and à la carte wins: an incremental à la carte dollar on a night you are already open costs you product, a little labor, and a card fee, and nothing else.

The correct conclusion is not "events are worse." It is this: events are not competing with your à la carte business. They are competing with an empty room. Sixty-five incremental à la carte covers on a Tuesday would be more profitable than the event — and they do not exist. That is the entire case, and it is strong enough that it does not need the exaggeration.

False claim two: "an event is found money." An event consumes selling time, coordination time, a room flip, kitchen capacity at a fixed hour, and — on any night you are open — à la carte covers you would otherwise have sold. Chapter 24 gave you the discipline: when you sell a seat-hour to one buyer, you have to charge the other buyer's price for it. Section 29.2 turns that into a dollar figure per night, and it is the number most independents never compute.

⚠️ Where the Money Leaks

Counting event revenue and forgetting event cost of sale.

Bellwether's fourteen events will consume, realistically, about 150 hours a year of the front-of-house partner's time: answering inquiries that never book, writing proposals, running site visits, building BEOs, revising BEOs, and standing in the room on the night. That is roughly three hours a week, every week, taken out of the person who is supposed to be running the floor.

Section 29.7 shows the event book producing about \$14,800 of contribution. Divided by 150 hours, that is about \$99 an hour of the FOH partner's time — which is a genuinely good rate for an independent operator's hour, and that is the honest test, not the gross revenue.

The leak is not selling events. The leak is selling them badly: chasing every inquiry, quoting from scratch each time, and burning ninety minutes on a proposal for a client who was always going to choose the hotel. Publish prices so unqualified inquiries filter themselves out (§29.9), use a template so a proposal takes twenty minutes, and protect the hour.

29.2 The private dining room, the buyout, and selling the space you already have

Bellwether has no private dining room. It is a 2,800-square-foot second-generation space with 1,700 square feet of front of house, a bar, and a wall of banquettes. There is no door to close.

Every consultant will tell you to build one. Do the arithmetic before you believe them.

Why Bellwether is not building a private dining room

A private dining room — a walled, doored space with its own service, sold separately from the main room — is the ideal instrument for this business, because it eliminates the one problem that makes events hard: it sells hours without taking them from anybody. You can run a 24-guest dinner behind a door on a Friday and not lose a single à la carte cover.

To create one at Bellwether, you would carve roughly 20 seats out of the 56-seat dining room. Here is what that costs.

The plan's Friday and Saturday do 120 and 123 covers. The kitchen's operating ceiling is about 132 covers — the hearth sustains 28 items an hour and 44% of entrées fire to order, so the fire, not the furniture, is what caps a big night. On those two nights the dining room is running at or near its practical limit already. Take 20 seats out and you have 36 dining seats plus 12 at the bar. Even at an aggressive 2.2 turns, 36 seats produce about 79 covers. Friday needs 120.

So a private dining room would cost Bellwether roughly 40 covers on each of its two best nights, permanently, in exchange for a room it might sell forty times a year. Forty Friday covers at a \$46 check is \$1,840 of revenue a night; across 52 Fridays and 52 Saturdays that is a six-figure hole, and the private room would have to produce more than that before it broke even. It will not.

The answer is no. Not "not yet" — no, on these numbers, in this room. That is a legitimate, defensible plan decision, and stating it plainly is worth more than a vague aspiration to "explore private dining."

What Bellwether has instead

What it has is a room that converts.

FIGURE 29.1 — The 40-top that already exists              [the Bellwether plan — schematic]

   NORMAL SERVICE                              EVENT CONFIGURATION
   ┌─────────────────────────────┐             ┌─────────────────────────────┐
   │  BANQUETTE RUN (long wall)  │             │  BANQUETTE RUN (long wall)  │
   │  [2][2][2][2][2][2]         │             │  ═══════════════════════════│
   │   [4]  [4]  [4]  [4]        │   ──flip──▶ │    ONE 40-SEAT TABLE        │
   │     [6]        [6]          │             │  ═══════════════════════════│
   │                             │             │                             │
   │  [2][2]   [4]  [4]  [4]     │  ← held     │  [2][2]   [4]  [4]  [4]     │
   │                             │    back     │                             │
   ├─────────────────────────────┤             ├─────────────────────────────┤
   │  BAR  ●●●●●●●●●●●●  (12)    │             │  BAR  ●●●●●●●●●●●●  (12)    │
   └─────────────────────────────┘             └─────────────────────────────┘
     17 tables · 56 dining seats                 5 tables · 16 dining seats
     + 12 bar = 68                               + 12 bar + one 40-top

   The banquette run — six two-tops (12), four four-tops (16), two six-tops (12) —
   is 40 seats that push together into a single long table. What remains for
   à la carte is two two-tops, three four-tops, and the bar.

   Not to scale. Legend: [n] = a table seating n · ● = a bar stool ·
   ═══ = banquettes and freestanding tables combined into one run.

This is the asset. It is not a private dining room and you should never sell it as one. It is a long table in a working restaurant, which is a genuinely appealing thing to sell to the right client — a retirement dinner, a rehearsal dinner, a company's end-of-quarter, a fiftieth birthday — and a terrible thing to sell to a client who wanted a boardroom.

Bellwether therefore has exactly three sellable event configurations, and you should be able to recite them:

* Standing capacity is not a seat count. It is set by the posted occupant load in your building and fire code, and you may not exceed it — see the compliance callout in §29.3.

The semi-private middle: a \$3,800 decision

There is one capital option worth pricing, because clients ask for it constantly: a ceiling track and drapery, plus a set of freestanding screens, that visually separates the banquette run from the rest of the room. Call it \$3,800 installed, illustrative.

It does not create privacy. It creates the impression of a defined space, which is enough to change what a client will pay and enough to convert a share of the inquiries that currently walk away. A defensible estimate: it converts about two additional partial-room events a year (about \$1,300 of contribution on the numbers in §29.7) and supports roughly \$200 more on the minimum for the five you were booking anyway (\$1,000 of revenue, about \$700 of contribution). Total ≈ \$2,000 a year against \$3,800 of capital — a payback a little under two years.

That is a real return. It is also entirely dependent on demand nobody has observed yet, because Bellwether has not opened. The plan's answer: not at opening. Revisit it after twelve months of actual event inquiries, when the two-year payback is based on a booking history instead of an assumption. Deferring a capital decision until the data exists is not indecision; it is the correct sequence, and it is exactly what Chapter 6 taught about spending money on a room before the room has told you anything.

👨‍🍳 On the Line

The room flip is labor, and nobody puts it on the event.

At 4:00 p.m. on an event Tuesday, somebody has to turn a banquette run into a 40-top. That means: pull four freestanding tables off the wall and reset them end-to-end, level them with shims because the floor in a 1920s warehouse building is not flat, clothe them (floor-length napery on an event, which means steaming the creases out because a folded rental linen looks like a folded rental linen), set 40 covers, place chairs, run the six-top ends, set glassware, place menus, and get the five remaining à la carte tables reset around the change.

It takes two people about seventy-five minutes if they have done it before and they have a diagram. It takes two people two hours the first time, and it takes three people and a raised voice if nobody wrote a setup diagram on the BEO.

Those are prep hours. They come out of the same day-shift bodies who were going to be breaking down chickens. That is why the event labor model in §29.7 carries an extra day prep shift, and it is why the BEO has a 4:00 p.m. flip line on it. If the flip is not on the schedule and not on the BEO, it happens at 5:15 in front of arriving guests, which is the visual equivalent of announcing that you do not do this often.

And then it flips back. At 10:30 p.m., after the last guest leaves, somebody strips 40 covers, breaks the run, and resets 17 tables for Wednesday lunch prep. Budget it. It is the single most commonly forgotten forty minutes in the event business.

29.3 Pricing per head: food and beverage minimums, site fees, and packages

There are three instruments, and most operators use only one of them. Use all three.

Per-person pricing is a fixed price per guest for a defined food package. It is what the client wants, because it is the only number that lets them budget. It is also what you want, because it converts an unknowable check average into a contractual one.

A food and beverage minimum is the least amount the client will spend on food and drink, regardless of how many people come or what they order. It is the instrument that protects capacity. Per-head pricing alone does not: a client who books your 40-top and then brings 22 people has bought your Tuesday night for 22 × \$58. A minimum says the night costs what the night costs.

A site fee (also called a room fee, or a rental fee) is a separate charge for the space and the setup, not for the food. It exists because some of what you are selling genuinely is not food: exclusivity, a room flip, an early access window, a coat check, extended hours. Charge it explicitly rather than burying it in the per-head price, for two reasons. First, it survives a client cutting their guest count. Second — and this matters — in many jurisdictions food and a site fee are treated differently for sales-tax purposes, and a fee is generally not part of the base that a service charge is calculated on. Keep them on separate lines.

Bellwether's published package sheet

The decision here is whether to publish prices at all. Many operators do not, on the theory that quoting individually lets you price to the client. In a 68-seat independent with no salesperson, publish. The unpublished price sheet costs you more in wasted proposals than it earns in occasional upsells, and the client who was never going to spend \$58 a head finds out on your website instead of on a Thursday afternoon in your dining room.

(All figures illustrative and constructed for the Bellwether plan.)

The package-versus-consumption decision is the single largest variable in event profitability, and it is worth being blunt about which way it cuts. On consumption, you sell what people drink at menu price and your pour cost lands where your pour cost lands. On a package, you have sold an all-you-can-drink window at a fixed price, and your cost is now a function of how thirsty the room is — which correlates with the occasion in ways you can predict but not control. A two-hour beer-and-wine package at \$30 a head is comfortably profitable at a corporate dinner and can be actively unprofitable at a wedding rehearsal where the average guest has four glasses of wine and the pour is being made by a bartender who is not counting.

Bellwether's default is on consumption with an authorized ceiling, and packages by exception. That is the right default for an independent that does not yet know its own event clientele.

Minimums, by night, built from what the night is worth

This is the table that makes the whole chapter operational, and every number in it is derived, not felt. The derivation is in §29.7; the logic is simply that the minimum has to clear the à la carte contribution the night would have produced on its own, plus the incremental cost of running the event, plus a margin for the disruption.

(Constructed for the Bellwether plan. Minimums are food and beverage before service charge and tax.)

Read the shape of that table rather than the individual numbers. The cheapest thing to sell is the hour nobody wants, and it gets progressively more expensive as it approaches Saturday at eight. The Monday minimum is 28% of the Saturday minimum for the same room and the same food, because on Monday you are selling a dark building and on Saturday you are selling somebody else's reservation.

Notice also that a partial-room Tuesday dinner at the \$2,400 minimum is a genuinely thin event — §29.7 puts it at \$713** of contribution after displacement, against **\$1,419 for the flagship Thursday event on a gross only 25% larger. A minimum is a floor, not a target. It is the number below which you say no. The plan's average event has to run near \$2,800–3,000, and the job of the person selling is to get there through the package tier, the beverage program, and the guest count — not by discounting to fill a Tuesday that was going to be quiet anyway.

🤝 Hospitality

Forty people who have never been here.

There is a second revenue line hiding in the event book and it does not appear on any event P&L.

A 40-guest private dinner puts forty people in your dining room, most of whom have never eaten your food. They did not choose you — somebody chose you for them — which means they arrive with no expectations to disappoint and a natural willingness to be impressed. Chapter 27 costs an acquired cover at something in the range of \$8–12 through paid channels for a restaurant of this size (illustrative; your number will differ). Forty guests is therefore on the order of \$320–480 of acquisition cost you did not spend, before a single one of them comes back on their own.

This has a practical consequence for how you run the night. The temptation with a private party is to treat it as a production problem — get forty plates out, clear, done. Treat it instead as forty first visits. The chef comes out for two minutes before dessert. The captain hands the host a card with the direct line for their next booking. Every guest leaves with something small that has the restaurant's name on it — the printed menu from their place setting is free and works fine.

The events that produce the second booking are almost never the ones where the food was best. They are the ones where somebody at that table felt like a regular for ninety minutes.

⚖️ Code and Compliance

Three rules that will bite a private-event program, and all three vary by jurisdiction.

Occupant load. Your maximum occupancy is a number set by the building and fire code and posted in the building — it is not your seat count, and for a standing reception it is the only number that matters. A 68-seat room may be permitted for considerably more standing, or considerably less than you assume, depending on egress, sprinklers, and assembly classification. Find the posted number, put it in the sales materials, and never sell past it. "We can probably squeeze them in" is how a restaurant ends up on the wrong side of a fire marshal.

Alcohol in a private-event configuration. A license permitting on-premise service to the general public does not automatically permit every arrangement a client will propose: a client-supplied case of wine, a cash bar at a ticketed event, a bar that stays open past your licensed hours, or service in a portion of the building not covered by your licensed premises diagram. Corkage, in particular, is permitted in some jurisdictions and flatly prohibited in others. Dram-shop exposure (Chapter 16) does not soften because the guest list was private — arguably the opposite, since an open-bar event is a structure that encourages over-service. Your service standards, ID checks, and refusal training apply identically.

Fee disclosure. A growing number of states and cities regulate how mandatory fees and service charges must be disclosed in food-service transactions, and some have specific rules for banquet contracts. §29.6 covers the substance. The procedural point: what your contract must say, and where it must say it, is jurisdiction-specific and changes.

Verify all three locally, and have an attorney read your event contract once. It is a few hundred dollars against a document you will use two hundred times.

29.4 The BEO: writing an event order that prevents arguments

Every bad event story you have ever heard has the same root cause. Somebody said something to somebody, and nobody wrote it down.

The banquet event order — BEO, and you will hear it pronounced as three letters — is the single document that carries an event from the sales conversation to the floor. It is the contract's operational twin: the contract governs money and liability, and the BEO governs what actually happens between 4:00 p.m. and midnight. Written properly, it is simultaneously the kitchen's production sheet, the bar's setup list, the floor's timeline, the office's invoice basis, and the client's written confirmation of every promise made.

Its real function is narrower and more valuable than any of that: the BEO exists so that nothing on the night has to be settled from memory. When the client's brother-in-law says at 8:15 that they were promised a second passed item, there is a document, the client signed it, and the conversation takes eleven seconds instead of ruining the evening.

The rules of a working BEO

1. One page front, one page back. If it is four pages, nobody on the line reads it.
2. Version it and date it. "Rev. 3 of 3, issued Oct 3." Two versions in circulation is the most common way an event goes wrong.
3. Numbers, never adjectives. Not "some passed items." Six passed items, three varieties, from 6:30 to 7:15.
4. Every time is a clock time, including the ones nobody thinks to write: room flip, client access, guest arrival, seating, each course, bar close, and the hard out.
5. Counts are on the document, broken out by choice — 24 chicken, 16 cabbage — because the chef orders from this and only this.
6. Allergies are on the document and on the chair, not on a list somebody is holding.
7. The distribution list is fixed: client, chef, bar, captain, office. Everyone gets the same revision at the same time.
8. It is signed. By the client and by the house. An unsigned BEO is a wish.

Here is Bellwether's, for the event this chapter costs in §29.7.

🧾 Read the Numbers

```text FIGURE 29.2 — "The banquet event order that prevents the argument" [the Bellwether plan — modeled] THE ARTIFACT Banquet event order #29-014, one page front and back, issued nine days out for a 40-guest plated private dinner. Signed by the client and the front-of-house partner; distributed to chef, bar, captain, and office. THE CONTEXT Bellwether, a Tuesday in October. Plan covers for a normal Tuesday are 62. The banquette run is configured as a 40-top; five à la carte tables and the bar continue to trade. The client is a professional-services firm marking a retirement; the billing contact and the day-of contact are two different people, which is the usual arrangement and the usual source of trouble.

┌────────────────────────────────────────────────────────────────────────────────┐ │ BELLWETHER — BANQUET EVENT ORDER BEO #29-014 │ ├────────────────────────────────────────────────────────────────────────────────┤ │ CLIENT [client organization] BILLING CONTACT [on file, w/ card] │ │ DAY-OF [name + mobile — on this page, not in someone's phone] │ │ EVENT Private dinner — retirement REV. 3 of 3 · FINAL ✓ │ │ DATE / DAY Tuesday, October 14 ISSUED Oct 3 SIGNED Oct 4 │ │ SPACE Banquette run, configured as one 40-seat table │ │ GUARANTEE 40 (due 12:00 noon Tue Oct 7 — received Oct 6) │ │ SET FOR 42 BILLED ON the guarantee or the actual count, whichever │ │ is higher │ ├────────────────────────────────────────────────────────────────────────────────┤ │ TIMELINE 4:00p ROOM FLIP begins (setup diagram, reverse side) │ │ 5:00p à la carte opens on 5 tables + bar; captain briefs floor │ │ 5:30p bar set · ice · glassware staged · menus placed │ │ 6:00p CLIENT ACCESS — place cards, one 6-ft gift table at the end │ │ 6:30p GUEST ARRIVAL · passed bites begin · bar opens │ │ 7:15p guests seated · first course fires │ │ 7:45p ENTRÉE COURSE FIRES — one batch (see KITCHEN NOTE) │ │ 8:45p dessert family-style · coffee service │ │ 9:30p bar closes │ │ 10:00p HARD OUT — contracted; overtime billed per contract §9 │ ├────────────────────────────────────────────────────────────────────────────────┤ │ FOOD — The Rivermill, plated three courses 40 guests @ $58.00/pp │ │ PASSED 6:30–7:15 hearth flatbread · smoked trout · squash arancini │ │ FIRST (guest choice — counts confirmed with guarantee) │ │ 26 × hearth greens, buttermilk 14 × squash soup │ │ ENTRÉE (counts confirmed Oct 6 — NO day-of changes) │ │ 24 × HEARTH CHICKEN 16 × charred cabbage + farro │ │ ALLERGY 2 × gluten-free (cards ON THE CHAIRS, seats 11 and 12) │ │ DESSERT family-style apple crumble + spiced cream — 5 platters │ │ Bread and cultured butter on the table at seating │ ├────────────────────────────────────────────────────────────────────────────────┤ │ BEVERAGE ON CONSUMPTION at menu price — no package │ │ Available: 4 wines by the glass · 3 drafts · full bar · NA punch │ │ CLIENT AUTHORIZES up to $900. BAR NOTIFIES CAPTAIN AT $750. │ │ Client-supplied wine: none. Corkage: n/a. │ ├────────────────────────────────────────────────────────────────────────────────┤ │ SETUP / VENDORS (included in the per-person price unless marked ‡) │ │ Floor-length napery, 40 covers ............ rental co., delivers 2:00p │ │ Coupes, 60 ................................ rental co., same delivery │ │ Low arrangements, 6 ....................... florist, delivers 5:00p │ │ Printed menus, 44 ......................... in house │ │ ‡ Client add-ons (AV, specialty rental, cake) rebilled at cost + 15% │ ├────────────────────────────────────────────────────────────────────────────────┤ │ KITCHEN NOTE Entrée course fires as ONE BATCH at 7:45p, ahead of the │ │ à la carte peak. Hearth is BLOCKED 7:30–8:15p. Ticket times │ │ on the 5 remaining tables will run long — captain warns those │ │ tables at seating and the kitchen holds two hearth slots. │ │ STAFFING 1 captain · 2 event servers · 1 event bartender │ │ +1 line cook · +1 prep (day) · +1 dish 4:00–7:30p │ ├────────────────────────────────────────────────────────────────────────────────┤ │ MONEY │ │ Estimated food and beverage ................................. $3,000.00 │ │ Service charge, 22% of F&B .................................. $ 660.00 │ │ MANDATORY. 18 points to the service team, 4 points retained by │ │ the house. THIS IS NOT A GRATUITY — see contract §6. │ │ Sales tax, 7.0% on F&B and service charge [illustrative rate] . $ 256.20 │ │ ───────────────────────────────────────────────────────────────────── │ │ ESTIMATED TOTAL ............................................. $3,916.20 │ │ Deposit received 8/19 (25% of minimum, non-refundable) ...... $ 600.00 │ │ Second payment received 9/14 (50% of minimum) ............... $1,200.00 │ │ BALANCE DUE at event close, card on file .................... $2,116.20 │ ├────────────────────────────────────────────────────────────────────────────────┤ │ SIGNED client __ Bellwether __ date ____ │ └────────────────────────────────────────────────────────────────────────────────┘

WHAT IT SHOWS A complete, executable night. Every person who touches this event can work from this page alone: the chef knows to order 24 birds and 16 cabbages and to fire one batch at 7:45; the bar knows the ceiling is $900 and to speak up at $750; the captain knows the room flips at 4:00 and the guests are out at 10:00; the office knows $1,800 is already collected and $2,116.20 clears on the card that night. The estimated total foots exactly: $3,000 + $660 + $256.20 = $3,916.20, less $1,800 received, leaves $2,116.20. WHAT IT DOESN'T It does not tell you whether the event makes money. There is not a single cost on this page — no food cost, no labor, no rental cost, and no charge for the twelve à la carte covers the 40-top displaces. A BEO is a production and billing document; the event P&L in §29.7 is a different document, and an operator who reads only the BEO will conclude that $3,916.20 walked in the door. It also does not settle what happens if the actual count is 34, or if the bar runs past $900, or if the party is still seated at 10:40 — those live in the contract, which is why the BEO points to it by section. THE DECISION Issue Rev. 3 as FINAL and freeze it. Distribute to all five parties the same afternoon. Any change after this point gets a new revision number and a fresh signature — no verbal amendments, no "the client mentioned." Put the setup diagram on the reverse so the 4:00 flip does not require a conversation. THE LESSON An event order is not paperwork. It is the mechanism that converts a promise made in a sales conversation into an instruction a cook can execute, and it is the only reason a disagreement on the night takes eleven seconds instead of costing you the room's goodwill and a $660 service charge. ```

Two details in that document deserve a second look, because they are the ones operators leave off.

The bar's notify threshold. The client authorized \$900 of beverage on consumption. The bar is instructed to tell the captain at \$750. That single line is what prevents the most common event argument in existence — the client who did not realize an open tab was open. At \$750 the captain goes to the host, quietly, and asks whether to keep going. The host says yes and now it is their decision, or says close it and you have protected the relationship. Either outcome is better than a surprise at the bottom of an invoice.

"Set for 42, billed on the guarantee or the actual, whichever is higher." You prepare a small overset — here 5% — because an empty place setting at a 40-guest dinner is conspicuous and a missing one is worse. You do not bill for the overset. You bill on the larger of the guaranteed count and the number who actually came. That asymmetry is the industry standard and it is fair in both directions: the client cannot shrink their commitment after you have bought the food, and cannot bring six extra people and expect not to pay for them.

29.5 Deposits, guarantees, and cancellation: getting paid for capacity you committed

An event is a sale of capacity, and capacity is perishable in exactly the way Chapter 24 described. When you accept a Friday banquette booking in August for October, you stop selling those seats. If the client cancels on October 5th, you do not get August back. You get four days to re-sell a Friday, which you will not do.

That is the entire justification for deposits and cancellation terms, and it is worth stating in those words to a client who bristles at them. You are not distrusting them. You are pricing the option you just handed them for free.

The three-payment structure

The BEO in §29.4 shows the structure Bellwether uses, and it is the simplest one that works:

For the \$2,400 minimum in the worked BEO that is \$600, then \$1,200, then a \$2,116.20 balance against a \$3,916.20 estimated total. **By the day of the event, \$1,800 of a \$3,000 food-and-beverage sale is already in the bank.** No à la carte cover in the building does that.

The last row matters more than it looks. Charge the balance the night of the event, on a card authorized in advance. Independents invoice, then chase, and a startling share of event revenue ages past sixty days for no reason other than that nobody wanted an awkward conversation at 10:15. Chapter 33 will show what receivables do to a business with a \$8,700 residual cash reserve; the short version is that an event you have not been paid for is worse than an event you did not book, because you have already spent the food cost.

The guaranteed count

A guaranteed count is the number the client is billed for regardless of who shows up. It is due at a stated hour — Bellwether uses noon, three business days out — and once given it cannot go down.

That single clause resolves the most common event dispute in the industry. Forty people confirm; thirty-four arrive; the client wants to pay for thirty-four. The kitchen bought forty, prepped forty, and staffed forty. The guarantee is what makes that conversation a matter of reading the contract rather than negotiating on the night in front of the client's guests.

The reciprocal obligation is real and you should honor it visibly: you set for more than the guarantee. The BEO says set for 42, billed on the guarantee or the actual count, whichever is higher. Two extra covers is a cheap insurance policy against the single worst event failure, which is a guest with no chair.

⚠️ Where the Money Leaks

The cancellation schedule nobody writes down.

Most independent event contracts say something like "deposits are non-refundable." That is not a schedule; it is a single blunt term that clients fight and that a restaurant frequently waives to avoid a public argument — which converts it into no term at all.

A schedule prices the thing being lost, which is re-sale time:

Cancelled	Client forfeits
More than 90 days out	The 25% booking deposit only
30–90 days	50% of the minimum
8–29 days	75% of the minimum
7 days or fewer	100% of the minimum

The logic is legible to a reasonable client: at ninety days you can re-sell a Friday; at six days you cannot. A term a client understands is a term you can actually enforce, and a term you enforce consistently is worth several times a harsher term you waive half the time.

One further clause, and it is the one that saves the December bookings: a date change is not a cancellation if the new date is within ninety days and is not a Friday or Saturday in November or December. Give the client somewhere to go and they will take it, which converts a total loss into a rescheduled event on a night you wanted to sell anyway.

29.6 Service charge versus gratuity: a distinction with teeth

This is the section where an event business most often creates a liability it does not know it has, and the distinction is genuinely simple even though its consequences are not.

• A gratuity is voluntary. The guest decides whether to leave it and how much. It belongs to the employees.
• A service charge is mandatory. It appears on the contract, the client cannot decline it, and — this is the part operators get wrong — under the federal framework Chapter 20 laid out, it is generally the employer's money in the first instance, not a tip.

That last point has three consequences that compound.

It is revenue, and it is wages when you distribute it. A mandatory service charge you pay out to staff is generally treated as wages rather than tips. It goes through payroll, it is subject to payroll taxes, and it counts toward the regular rate for overtime purposes. It is not eligible for a tip credit, because it is not a tip.

It changes the overtime arithmetic. Chapter 19 built a labor model against 453.5 hourly hours; Chapter 20 showed how quietly an exempt classification can fail. A service charge distributed as wages raises the regular rate for the week in which it is earned, which raises the overtime premium for anyone over forty hours in that week. An event on a Saturday, at the end of a heavy week, is exactly when that bites.

It must be disclosed clearly. Several jurisdictions have specific rules about how a mandatory charge is described on a menu, a contract, or a check — precisely because guests reasonably assume a "service charge" goes to the server. Describing a charge in a way that implies it is a gratuity when it is not is the kind of thing that produces both a regulatory problem and a genuinely angry guest.

⚖️ Code and Compliance

What you must actually do, and what you must verify locally.

The framework is federal; the specifics are not. Verify all of this with an employment attorney licensed where you operate before you print a contract.

1. Say which it is, in words, on the contract and on the final bill. Bellwether's BEO reads "22% service charge (mandatory; distributed to the event service team as wages)." Nine words that prevent a great deal of trouble.
2. If you distribute it, run it through payroll. Not the tip pool, not cash, not "we hand it to the captain." Payroll — because it is wages, and because Chapter 34's controls chapter will want the audit trail.
3. Do not take a tip credit against it. It is not a tip.
4. Recompute overtime in the week it is earned. A distributed service charge raises the regular rate.
5. Decide, in advance and in writing, whether the client may add a voluntary gratuity on top — and make clear that any such gratuity is separate, voluntary, and goes entirely to staff.
6. Some jurisdictions require specific disclosure language or restrict the term "service charge" entirely. Several have moved on this recently. Check yours; do not copy a contract from a restaurant in another state.

The safest posture for a small independent is the one Bellwether takes: a mandatory service charge, disclosed as mandatory, distributed as wages through payroll, with no tip credit taken against it, and a clearly optional gratuity line left blank.

Notice what the 22% actually buys the restaurant, because this is the commercial logic underneath the compliance: on the worked \$3,000 event it produces **\$660, and §29.7 shows direct event labor at \$692. The service charge very nearly pays the event's labor.** That is its function. It is not margin, and an operator who reads it as margin will price events as though labor were free and then wonder why a busy event season produced no money.

29.7 Costing an event to a true contribution margin

Here is the arithmetic the whole chapter has been walking toward, and it contains the number most event businesses never compute: displacement.

An à la carte cover costs you food, a share of labor, and nothing else — the room was going to be open anyway. An event costs you food, labor, rentals, coordination time, and the covers you could not sell because the event was sitting in them. Leave displacement out and every event looks wonderful.

🧮 Run the Numbers

The flagship: a Thursday banquette 40-top at \$3,000.

Forty guests, \$75 a head — \$52 food, \$23 beverage. Plated, three courses, wine on consumption to a \$900 ceiling.

	Amount
Food and beverage revenue	\$3,000.00
Food cost (27% of \$2,080) | −\$561.60
Beverage cost (22% of \$920) | −\$202.40
Total COGS (25.5%)	−\$764.00
Direct event labor \$692, less the \$660 service charge that funds it	−\$32.00
Rentals — linen, chair covers, the AV the client asked for	−\$180.00
Incremental cleaning and room flip	−\$95.00
Coordination and selling time (3.5 salaried hrs @ \$32) | −\$112.00
Breakage and misc allowance (1%)	−\$30.00
Displacement: 20 covers × \$46 × 40% contribution | **−\$368.00**
CONTRIBUTION	\$1,419.00

Check: \$764 + \$32 + \$180 + \$95 + \$112 + \$30 + \$368 = \$1,581; \$3,000 − \$1,581 = \$1,419, or 47.3% of the gross.

Now the thin one: a partial-room Tuesday at the \$2,400 minimum.** Thirty guests, \$80 a head. Same room set-up, same coordination effort, same rental order — because effort does not scale with guest count.

	Amount
Food and beverage revenue	\$2,400.00
Total COGS (25.5%)	−\$612.00
Direct event labor \$640, less the \$528 service charge	−\$112.00
Rentals	−\$180.00
Incremental cleaning and room flip	−\$95.00
Coordination and selling time (the same 3.5 hours)	−\$112.00
Breakage and misc (1%)	−\$24.00
Displacement: the banquette run removes 32 of 56 dining seats on a 62-cover night — 30 covers × \$46 × 40% | **−\$552.00**
CONTRIBUTION	\$713.00

Check: \$612 + \$112 + \$180 + \$95 + \$112 + \$24 + \$552 = \$1,687; \$2,400 − \$1,687 = \$713, or 29.7%.

Read the two together. The gross differs by 25%. The contribution differs by 99%. Three things did that: the fixed coordination cost fell on a smaller sale, the per-head price bought fewer heads, and the displacement was larger on the quieter night — which is the opposite of what intuition says.

That last point is the one worth carrying out of this chapter. Tuesday feels like the safe night to sell because the room is empty. But the banquette run is the dining room on Tuesday; taking it removes half the seats on a night with only 62 covers to lose. On Thursday you displace a fifth of the room. On Tuesday you displace half of it.

Buffet versus plated

The received wisdom is that buffets are cheaper. They are cheaper in labor and more expensive in food, and which wins depends entirely on the guest count.

A buffet must not run out, which means you produce for roughly 115–120% of the guaranteed count on the categories guests take freely. Call it fifteen points of extra food. Against that, a buffet at forty guests needs perhaps two service staff instead of three plus a captain running courses — call it four to five labor hours saved.

At Bellwether's numbers, on the \$3,000 event: fifteen points of extra food on \$2,080 is about \$312**; five hours of saved service labor at \$18 is \$90**. **Plated wins by roughly \$222 at this size. The crossover comes when guest counts get high enough that the labor saving scales and the food overage does not — somewhere north of eighty guests for most independents, which is a size Bellwether's room cannot hold. For a 68-seat restaurant, plated is almost always the right answer**, and the reason is arithmetic rather than aesthetics.

⚠️ Where the Money Leaks

Three costs that never make it onto the event P&L.

The coordination hour. Every event in the two worked examples above carries \$112 of salaried time — three and a half hours of somebody's week spent on menus, seating, dietary restrictions, and four emails about parking. It is identical for a \$2,400 event and a \$9,000 one, which is the strongest argument in this chapter for a minimum. Fourteen events is roughly forty-nine hours a year, or a week and a quarter of a manager's life, and it appears on no schedule.

The Monday after. An event ends at ten and the room has to be reset, the rentals stacked for pickup, and the linen counted. The \$95 in the models is generous for a banquette 40-top and light for a buyout.

The service you did not run. This is displacement, and it is the big one — \$368 on Thursday, \$552 on Tuesday. An operator who omits it will conclude events run at 60% contribution and price them accordingly, which is precisely how a restaurant ends up busy, tired, and no better off.

29.8 Off-site catering: what changes when the food leaves the building

Off-site catering looks like an extension of the event business and is closer to a different business. Three things change, and all three cost money.

You lose the kitchen. Everything must be produced in advance, transported at temperature, and finished with whatever equipment fits in a van. Bellwether's food is built around a wood-fired hearth — the single least portable piece of equipment in the building. A menu that travels is a different menu, and pretending otherwise produces exactly the failure Chapter 28 identified for delivery: the dish arrives, and it is not the dish.

You take the food-safety risk with you. Chapter 25 owns the framework; what matters here is that time and temperature control does not become optional because the food is in a vehicle. Hot food must stay at or above 135°F and cold at or below 41°F in transit, holding equipment must actually hold, and somebody must be logging temperatures at departure and arrival. An off-site event is also frequently outside your own health jurisdiction and may require a separate temporary permit — verify locally, every time, and do not assume your establishment permit travels.

You inherit somebody else's building. Power, water, access, parking, elevator, and whether there is anywhere to wash a hand. The site visit is not optional and it is not free; add it to the coordination hours.

The honest recommendation for a restaurant Bellwether's size, in year one: do not chase off-site catering. Take the one that walks in from a good regular, price it at a premium that reflects the real cost, and treat it as a relationship investment rather than a revenue line. The plan's \$42,000 does not depend on it, and a young restaurant with a four-person line and a manager already spending forty-nine hours a year on coordination does not have the capacity to build a second operation.

👨‍🍳 On the Line

What actually goes wrong off-site, in order of frequency.

Not food poisoning. The realistic failure list, from operators who do this regularly:

1. Something was left at the restaurant. The single most common off-site failure, and the reason a written load-out list exists and gets signed by two people.
2. The power was not what was promised. Two chafers and a coffee urn on one household circuit is a dark room and a cold entrée.
3. The room was not ready when you arrived, and forty-five minutes of your load-in window evaporated.
4. Nowhere to put dirty plates, because nobody asked.
5. Holding time ran long because the client's programme ran long, and the food that was perfect at 6:30 was served at 7:40.

Every one of those is prevented by a site visit and a load-out list, and every one of them costs more than the site visit would have.

29.9 Selling fourteen events: the calendar, the accounts, and the inbound you already have

Fourteen events in a year is roughly one every twenty-six days. That is a small number, and it changes what selling means: you are not building a pipeline, you are converting the inquiries you already receive and cultivating perhaps four or five repeating accounts.

Inbound first, because it is free. A restaurant with a private-events page, a published minimum, and a form that reaches a human within one business day converts a meaningful share of the people who already wanted to book something. Chapter 27 owns the marketing mechanics; the event-specific part is that publishing the minimum is a filter, not a deterrent. The client who leaves because the Friday minimum is \$4,000 was never going to spend \$4,000, and you have saved yourself the coordination hours.

Then the four accounts that repeat. Corporate holiday parties, a professional-services firm's client dinners, a neighborhood association, a family that does the same birthday every year. Repeating accounts are worth disproportionately more than their revenue, because the coordination cost collapses on the second booking — the menu conversation has already happened.

The calendar is the strategy. Sell the hours the dining room cannot: Monday and Tuesday first, then Wednesday and Thursday, then Friday, and treat Saturday as something you sell only at a price that beats a full service. December is a different market — it is the one month a neighborhood restaurant can command a genuine premium, and the December minimums in §29.3 reflect that. Book December in September.

🤝 Hospitality

The event guest is forty first visits.

Chapter 23 established that a first visit is expensive to acquire and a second costs nothing, and that Bellwether's plan needs 9,035 guests coming four times a year. An event puts forty people in your dining room at once, most of whom have never been, all of whom are in a good mood, and none of whom paid — because the host did.

That is the most efficient acquisition event available to a restaurant, and it is almost never treated as one. The cost is roughly zero: a card at each place setting, a word from the captain about the Tuesday brunch, a menu they can take. If two of those forty guests become four-visit regulars, Chapter 23's arithmetic says that is \$441.60 of lifetime contribution — a third of the entire event's margin, earned by an act of attention that cost nothing.

Price the event on the event. Then remember you also just met forty people.

🍽️ The Business Plan

Checkpoint 29 of 40 — the Catering & Events section.

The verdict on the \$42,000: KEPT — with the contribution rate corrected downward.

The plan's fourteen events at an average \$3,000 are achievable in this room, on this calendar, at these minimums. The composition that produces it:

What must be corrected is the margin, not the revenue. A plan that assumes events contribute at the restaurant's blended 40% would book \$16,800 of contribution. The honest figure, after displacement, is:

• 8 thin weeknight events at ~\$713 = **\$5,704**
• 6 stronger Thursday/Friday events at ~\$1,419 = **\$8,514**
• Total contribution ≈ \$14,218 — 33.9% of event revenue, not 40%.

The \$2,582 difference is displacement, and it is the single thing an event business must not forget to count.

What this checkpoint adds to the plan:

1. The minimum and site-fee schedule from §29.3, published on the events page.
2. The BEO template from §29.4 — the one document that prevents the arguments.
3. The three-payment structure and the cancellation schedule from §29.5, including the date-change clause that converts cancellations into reschedules.
4. A service-charge policy — 22%, mandatory, disclosed as mandatory, distributed as wages through payroll, no tip credit taken against it, voluntary gratuity line left blank and clearly optional.
5. A displacement rule: no event is quoted without subtracting the covers it occupies, at \$46 and 40%.
6. A calendar rule: Monday through Thursday first; Friday at the banquette minimum; Saturday only at a price that beats a full service; December booked by September.

What this checkpoint does not settle. Whether fourteen inquiries actually arrive in year one — the plan assumes a conversion rate on inbound demand that a restaurant with no operating history cannot evidence. Chapter 27's marketing plan carries the events page but no dedicated event marketing spend, which is deliberate and may prove wrong. And the forty-nine annual coordination hours sit inside a salaried manager's week that Chapter 19 already showed is fully committed.

Open questions carried forward:

1. If events run at 33.9% rather than 40%, what does that do to the plan's operating profit? (Chapter 31)
2. Fourteen events cost forty-nine coordination hours. Whose hours, and what came off their plate? (Chapters 19 and 21)
3. What happens to the \$42,000 if the first December is soft? (Chapters 33 and 39)

Conclusion

An event is a sale of capacity, and capacity is the one thing a restaurant cannot manufacture. That is why events work: they let you sell Tuesday at Thursday's price, they hand you known covers and a known menu, and they put the money in the bank before the food is bought.

It is also why the arithmetic is more hostile than it looks. Every event occupies seats that could have been sold, and on the quiet nights that feel safest to sell, the displacement is largest — half the dining room on a Tuesday against a fifth of it on a Thursday. A restaurant that prices events without subtracting displacement will conclude they run at sixty percent contribution, book as many as it can get, and discover at the end of a busy year that it was tired and no richer.

Bellwether's fourteen events survive that test. They contribute about \$14,218**, which is real money on a plan projecting \$261,020 of operating profit — but it is 33.9%**, not the 40% a careless model would assume, and the difference is \$2,582 that only appears if somebody counts the empty chairs.

Write the BEO. Publish the minimum. Take the deposit. Charge the card the same night. And when forty people you have never met are sitting in your dining room in a good mood, remember that the host paid for all of them and none of them owe you anything yet.

Chapter 30 looks at the other direction entirely — the formats that need no dining room at all.

Key Terms

Banquet event order (BEO) — the single-page operating document that specifies an event's timing, counts, menu, service, beverage authority, and billing. The document that prevents essentially every argument events generate. (Ch. 29)

Per-person pricing — quoting an event as a price per guest rather than a total, so the cost moves with the count and the guarantee governs the floor. (Ch. 29)

Food and beverage minimum — the minimum spend on food and beverage a client commits to for a given room, day, and time. A floor below which you decline, not a target. (Ch. 29)

Site fee — a separate charge for the space itself, distinct from the food-and-beverage minimum; it prices exclusivity rather than consumption. (Ch. 29)

Event deposit — money taken at signing and at intervals thereafter to price the capacity the client has removed from sale. Bellwether uses 25% non-refundable at signing, 50% at thirty days, balance on the night. (Ch. 29)

Cancellation terms — the schedule stating what a client forfeits at each distance from the event date, priced against how much re-sale time remains. (Ch. 29)

Guaranteed count — the number the client is billed for regardless of attendance, due at a stated hour and unable to be reduced thereafter. The reciprocal obligation is that the restaurant sets for more. (Ch. 29)

Buyout — the purchase of an entire service, closing the restaurant to the public. Priced against what the displaced service would have produced, not against the food. (Ch. 29)

Buffet vs. plated cost — buffets trade higher food cost (production at 115–120% of the guarantee) for lower labor; at counts under roughly eighty guests, plated generally wins. (Ch. 29)

Event labor model — the direct staffing of an event — captain, servers, bartender, kitchen — costed separately from the restaurant's schedule and largely funded by the service charge. (Ch. 29)

Off-site catering — production for service away from the licensed premises, which removes the kitchen, carries the food-safety obligation into transit, and frequently requires a separate temporary permit. (Ch. 29)

Displacement — the contribution forgone from covers an event occupies. The cost most event businesses never compute, and the one that separates a 47% event from a 30% one. (Ch. 29)

Spaced Review

1. Without looking back: state the three things an event gives a restaurant that à la carte service does not, and the one claim about event margin this chapter says is false.
2. A Tuesday event grosses \$2,400 and a Thursday event grosses \$3,000 — a 25% difference. Why is the difference in contribution 99%? Name all three causes.
3. From Chapter 24: Bellwether's room runs 43% empty on Saturday while the hearth runs over rate. Explain why that fact argues against selling Saturday buyouts cheaply, using the displacement method from §29.7.
4. From Chapter 20: a mandatory service charge is distributed to the event team. State three consequences that follow, and one thing you must not do.
5. The recurring question: does adding a fifteenth event to the calendar move prime cost, and in which direction? What would you need to know about which night it lands on before answering?