Ch23 Discussion
Discussion Guide
1. "Only one of these is defensible." Which of your restaurant's assets could a competent competitor across the street take from you inside two years — and which could they not?
What to listen for: students should be able to run the list from §23.1 (menu, room, price, chef, sequence of service, suppliers) and land on the accumulated preference of a neighborhood as the one thing that is not transferable. Push back on anyone who names "our recipes" or "our concept" — both are copyable, and saying so out loud is the point. The strong answer connects defensibility to where the asset physically lives: in staff, over time, which makes it a Chapter 21 problem.
2. Chapter 14 priced the second Friday in October at $116.70 and said everything that mattered was off the P&L. This chapter says it may have cost $640.32. Which number would you put in front of a lender, and why?
What to listen for: the honest answer is $116.70, because it is the only one that is countable — and that is exactly the problem the chapter exists to name. Listen for students who recognize that the $640.32 rests on an attrition rate nobody can measure, and then for the better students who notice that the break-even ($77 must save 0.35 of a guest) requires no such assumption and is therefore the defensible thing to argue. Someone will say "so the real number is unknowable" — correct, and it is still a decision you have to make tonight.
3. Is a comp a cost or an investment? Take a position and defend it with the accounting.
What to listen for: the productive disagreement. On the P&L a comp is unambiguously a reduction in net sales that also distorts food cost percentage (§23.4's leak callout) — so as accounting, it is a cost. Judged against GLV with a 13.1% break-even, it is an investment. The best answers hold both: it is an investment line recorded in a cost account, which is precisely why it must be coded by reason and read weekly rather than judged by whether it came in under budget. Ask the follow-up that separates the room: what would convince you the comp line was too low?
4. §23.5 says recognition is welcome when it serves the guest and unsettling the instant it serves you. Test that rule against three hard cases: (a) the bartender who starts the regular's drink without asking; (b) the host who says "welcome back" to someone who has not been in for two years; (c) a note recording that a guest mentioned a difficult year.
What to listen for: (a) is a good case for asking rather than assuming — "the rye Manhattan again, or something different?" preserves the guest's agency and costs two seconds. (b) is usually fine, and the interesting version is what happens when the greeter is wrong. (c) should split the room, and the chapter takes a firm position: some things a guest tells you are for the human being they told, and writing them down is a small betrayal even when nothing bad comes of it. Do not resolve it too quickly; the discomfort is where the policy gets written.
5. The chapter refuses to tell you what a one-star increase is worth, and gives you a 0.72% break-even instead. Is that intellectual honesty or is it dodging a question the reader needs answered?
What to listen for: students who connect it to Chapter 1's treatment of the 90% failure myth have understood the book. Push the strongest objection — an operator has to make a budget decision and "we don't know" is not a budget — and then let them find the answer the chapter already gave: you don't need the effect size, you need to know whether the effect exceeds five and a half covers a week. This is the most transferable single idea in the chapter and it is worth ten extra minutes.
6. Chapter 21's retention bundle cost $19,716 and came out break-even on turnover alone. This chapter claims a 72% return on top of it at a tenth of a visit — and then says you cannot prove it. What would you actually do with an argument you believe and cannot demonstrate?
What to listen for: the honest operator's position — track both ends (average server tenure, repeat-visit rate), look at them side by side quarterly for two years, and accept that it is a comparison rather than a finding. Listen for students who reach for a randomized test and then discover a 68-seat restaurant has no control group. Then ask the uncomfortable follow-up: if you cannot prove it, how do you defend the line item when the business gets tight in February? There is no clean answer, which is the most useful thing about the question.
7. You have $5,000 of discretionary money and one quarter. Spend it on (a) a promotion that brings in new guests, (b) staff retention, or (c) recovery capacity on the floor. Defend your allocation in dollars.
What to listen for: whether they use the chapter's own arithmetic rather than intuition. Reference points: acquisition at roughly $87 per genuinely new guest (about 57 guests, $12,586 of three-year value, none of it this year); $5,000 of recovery capacity breaking even at 23 guests saved ($5,000 ÷ $220.80); the frequency lever at $14,108 per tenth of a visit. Any allocation is defensible; an allocation without a break-even attached is not. The strongest answers notice that (a) is the only option whose payback is measured in years.