Chapter 10 — Teaching Notes

What this chapter is doing

Chapter 10 is where the book's central claim gets its hardest test, because it is the chapter students arrive at wanting to talk about food.

The claim is that a menu is not a culinary document. It is a purchasing specification and a labor model wearing a description of dinner, and it sets both halves of prime cost before anyone is hired. Students who have worked in kitchens will resist this — not intellectually, but emotionally, because the menu is the part of the business they got into the business for.

Do not argue them out of it. Let the arithmetic do it. The chapter is built so that three successive calculations make the point without anyone having to be told: the \$3,020 hidden cost of one item, the 68% peak fire utilization, and the \$33.12 reconciliation. A student who works all three has changed their mind about what a menu is, and they did it themselves.

The second job is narrower and more transferable: §10.6 is the book's best teaching opportunity on evidence. It is a section where the correct professional posture is "this is widely believed and not established, and here is what I would do anyway." That is a hard posture to teach and a valuable one, and it pairs directly with Chapter 1's case study on the ninety-percent myth.

Timing

• A 75-minute session: §10.3 cross-utilization, worked live on the board with the poultry program (20 min) → §10.4 the \$3,020 calculation and the throughput math (30 min) → §10.7's reconciliation (20 min) → close on §10.6's evidence problem (5 min, as a provocation for reading). Assign §10.1, §10.2, §10.5, and §10.8 as reading.
• Two sessions: split after §10.4. Session one is the menu as an operating document (§10.1–10.5, the arithmetic). Session two is the menu as a sales document (§10.6–10.8), and it carries the harder discussion prompts, because the evidence material needs the trust that session one builds.
• Self-paced: 5–7 hours including exercises. Exercises 10.16–10.27 are the computational core.

Common misconceptions

1. "Cross-utilization means using cheap ingredients." The most frequent misreading, and it comes from students hearing "lower food cost" and filling in the mechanism themselves. The correction is Figure 10.3, worked live: Bellwether buys whole air-chilled chickens, which is not the cheap option, and gets a low food cost by reaching five menu items with them. Cheap food is how you lower food cost badly. Cross-utilization is how you lower it while buying well.

2. "A short menu means limited choice, and guests want choice." Students conflate item count with perceived variety. Show them Bellwether's menu and ask how limited it feels — twenty-two items across five techniques, two proteins they can share, and a vegetarian entrée in the main panel. Then show them a 41-item menu and ask which one sounds like the kitchen knows what it is doing. Perceived variety comes from range and technique, not from line count.

3. "The menu is written and then the kitchen is designed to make it." Almost always backwards in practice, and the chapter is explicit about the two-way dependency. §10.4's twelve-position hearth is the menu stating a requirement; if Chapter 7's equipment cannot meet it, one of the two documents has to change. Students find this genuinely clarifying — it is the first time in the book they see two chapters constrain each other rather than one feeding the next.

4. "Menu psychology is a science." They have read the golden-triangle advice somewhere, probably more than once, and often in a source that seemed authoritative. Handle this carefully: the point is not that the advice is wrong, it is that it is unverified, and that a professional should be able to say which of their practices are evidence-based and which are conventions they follow because they cost nothing. Watch for the student who over-corrects into "so none of it matters" — that is the wrong lesson, and §10.6's "design so you win either way" is the right one.

5. Confusing a standardized recipe with a cost card. Students merge them because both list ingredients. Put two boxes on the board and refuse to let them merge, exactly as Chapter 6 did with the contingency and the reserve. The recipe has quantities and no money; the cost card has quantities and money. The recipe is enforced by a scale; the cost card is enforced by a spreadsheet.

6. "The specials board is where creativity lives, so don't bureaucratize it." Sympathetic and expensive. The §10.8 callout is the corrective, and the number that lands is 156 uncosted items a year, disproportionately built from the most expensive proteins in the building. The reframe that wins students over is the R&D-pipeline argument: costing a special does not kill creativity, it turns the special into next spring's menu item.

7. "0.90 attachment means 90% of guests order an entrée." A subtle one worth catching, because it recurs in Chapters 12, 15, and 24. Attachment rates are averages across all covers and can exceed 1.00. The reason Bellwether's is 0.90 rather than 1.00 is the bar-first guest who orders two starters and no entrée — table 3 in the §10.7 sample.

The hardest point to teach

That the reconciliation in §10.7 is a constraint, not a result.

Students read the three tables that produce \$33.12 and conclude that the author was clever. What actually happened is the reverse: Chapter 2 fixed \$33.12 eight chapters earlier, and the menu had to be built to produce it. The prices are not free.

This is the single most valuable idea in the chapter for anyone who will write a business plan, and it generalizes far beyond menus. In a plan, most numbers are downstream of a number somebody already committed to, and the job is to find out which one. A student who understands this stops treating a business plan as a forecast and starts treating it as a system of simultaneous constraints.

The demonstration that lands: put \$33.12 on the board, alone. Then ask the room to write an eight-item entrée list that produces it, given attachment rates of 0.90 / 0.45 / 0.15. Give them twelve minutes. Most will fail on the first attempt, and the failure is the lesson — they will have written a menu they like and then discovered it produces \$27, or \$41. Then show them Figure 10.6 and ask what had to be true about the ladder.

A demonstration that works

The whiteboard third column.

Hand the room a 40-item menu — any real one, from anywhere, no framing. Ask them to make three columns: Item · Station · What else uses it.

Give them fifteen minutes and no other instruction. They will fill the first column instantly, the second with some argument, and the third barely at all — because the information is not on the menu and often does not exist in the restaurant either.

Then ask: "How many items have a blank third column?" On a typical long menu it is more than half. Every one of those is an ingredient family feeding a single dish.

The extension, if you have time: ask them to price the blanks using §10.4's model at their own estimate of a fully loaded wage. A room that discovers a menu is carrying \$40,000 of invisible annual cost has learned something no lecture delivers.

A second, faster demonstration: the placement-rate calculation from Case Study 2, discussion question 4. Bellwether's salsa verde appears on 3 of 22 dinner items — but on 2 of 8 entrées, and the two carry 35% of entrée orders (chicken 25% + hanger 10%). A two-top ordering two entrées meets it twice about $0.35^2 = 12\%$ of the time. Ask whether that is acceptable. There is no right answer and the argument is excellent.

Assessment notes

• Exercises 10.16, 10.19, 10.21, and 10.22 are the computational core of the check-build material. A student who can do those four can reconcile any menu to any target check, which is a genuinely employable skill.
• Exercise 10.22 (build a ladder to exactly \$26.00) is the best single diagnostic in the set. It is under-determined on purpose — there are many valid answers — so grade the shape of the ladder and the internal checks (high/low ratio, median vs. weighted average, where the heaviest share sits), not just the arithmetic.
• Exercise 10.24 is the "cost this" item. Watch for students who add ingredient cost; the whole point is that this figure exists before ingredients, and a student who mixes them has not understood the division of labor between Chapters 10 and 11.
• Exercise 10.26 is the staffing-allocation problem and it has two defensible answers (hearth vs. garde manger). Grade the reasoning. The arithmetic favors the hearth; the guest-experience argument favors garde manger; a student who names both and picks one with a stated criterion has done better work than one who simply picks the "right" answer.
• Exercise 10.27 is the best stress-test item. Students who get 88% and stop have done half the work; the question asks what they would change, and the good answers identify the flatbread as the only discretionary item colliding with entrée fire.
• Exercise 10.28 ("read this menu and find the leaks") is the diagnostic. Look for at least: 41 items on a three-person line; 11 proteins; lunch sandwiches printed on the dinner menu; one fryer serving multiple items; no evident cross-utilization; and — the one most students miss — that six entrées and four salads is a category imbalance that tells you nobody has ever asked what the restaurant is for.
• Exercise 10.33 is the ethics and judgment item. Part (d) — the honest refusal — reliably produces the better discussion, because students want to write a disclaimer and the assignment asks for a sentence a human would say. Grade on whether the "no" ends with a better "yes."
• Exercise 10.34 is the Business Plan extension and works well as a graded portfolio piece across two weeks. Require that every column resolve. A menu whose reconciliation is off by eleven cents has not met the standard, and saying so once early saves an enormous amount of argument later.

Connections forward

Flag these explicitly so students build the map:

• Chapter 11 costs everything written here. Teach 10 and 11 as a pair if you can; the recipe card in Figure 10.5 becomes Chapter 11's cost card line for line, and the Hearth Chicken is the hinge.
• Chapter 12 replaces every assumed order share in this chapter with measured point-of-sale data. The §10.7 sample Thursday — 32% chicken against a 25% assumption — is planted specifically to set up Chapter 12's insistence on sample size.
• Chapter 13 builds the order guide from Figure 10.3, and the single-supplier failure in Case Study 2 is its receiving lesson in advance.
• Chapter 14 runs the service that §10.4 said was executable, and measures the ticket times the recipe card asserted.
• Chapters 15 and 16 build the \$12.88 of beverage that sits beside the \$33.12 of food.
• Chapter 18 takes over §10.6's conclusion: the thing that reliably sells food is a server who can describe it.
• Chapter 24 measures the "55% of covers in a 90-minute window" that §10.4 assumed.
• Chapter 28 answers the question this chapter deliberately left open: which of these items travel?