Chapter 13 — Key Takeaways
The core claims
-
Food cost is not set at the stove. It is set on the order guide, at the receiving door, on the walk-in shelf, and on the count sheet. A cook who portions perfectly out of a badly bought, carelessly received, poorly stored walk-in produces a beautifully executed 34% food cost.
-
Buy programs, not dishes. Bellwether's thirty-four menu lines come from eight purchasing programs across fifty-four order-guide lines. Thirty-eight of those lines — 70% of the guide — carry 67% of the money. Cross-utilization does not just simplify a menu; it concentrates a purchasing system into a number of decisions one person can manage well.
-
A par level is a calculation, not a feeling. Forecast usage until the next delivery, plus a safety stock sized to what running out would cost. Run it forward through a full cycle; if you do not land on the same shelf you started from, the par is wrong.
-
Cross-utilization changes the correct size of a safety stock. On a line with a second outlet, the cushion gets sold at full margin two days later. On a line with one outlet, it is money you might throw away. That is why Bellwether can carry eight spare halves into a Saturday night.
-
Receiving is the only moment you can still say no — and it returns about 3.4× on the manager hour it costs. Twenty minutes a morning, five mornings a week, is \$1,907 of labor against roughly \$6,454 of short counts, light weights, substitutions, and price creep.
-
Get the credit memo at the door, signed, and verify it lands on the statement. A verbal "I'll take care of it" is worth nothing, and a credit issued and never applied is invisible money.
-
The walk-in is a financial statement. The same neglect that puts raw chicken above ready-to-eat greens and lets a box drift to 46°F produces both a health inspector's report and a food-cost variance. Storage order, labeling, dating, and FIFO are one discipline serving two masters. (The inspection itself, HACCP, and the Food Code belong to Chapter 25.)
-
Without an ending count you do not have a food cost. You have an opinion with a percent sign after it. Count in shelf order, two people, same day, same time, every period — because a consistent counting bias is a rounding error and an inconsistent one is noise that destroys the instrument.
-
Inventory is cash you decided to store as food. Turnover is the bridge between food cost and the bank account, and it has an optimum rather than a maximum — push it too far and you are 86ing the signature dish on a Saturday.
-
The variance investigation order is unchanged from Chapter 11: theft is seventh. What this chapter adds is that most of the causes ahead of it are physical — fixable with a scale, a thermometer, a label gun, and a shelf plan, without a conversation about anybody's honesty.
The formulas
PAR = (forecast usage per service x services until next delivery) + safety stock
USAGE (Ch. 11's arithmetic, this chapter's procedure)
= beginning inventory + purchases - ending inventory
... adjusted for transfers between food and beverage, and credits applied
FOOD COST % = food used / food sales
SHORTCUT ERROR = (change in inventory) / sales <- exactly the gap between
"invoices over sales" and the truth
INVENTORY TURNOVER = cost of product used / average inventory
DAYS ON HAND = days in period / turnover for that period
The rules of thumb
| Prime-vendor strategy | Commit the commodity. Shop the identity. |
| Food inventory, full service | 4–8 days on hand ≈ 45–90 turns a year (orientation, not a target) |
| Receiving | 20 minutes a morning, a named person, on the printed schedule |
| Counting cadence | key items daily (12 lines, ~55% of the money, ~12 min) · full food weekly · everything monthly |
| Waste log | at or below 1.0% of food sales; the log is a floor, the variance is a ceiling |
| Specs | write one for anything expensive, variable, or load-bearing. Twelve, not fifty-four. |
| Second source | live, not listed — one order a quarter, for every program reaching 4+ menu items |
The Bellwether numbers to remember
- Order guide: 54 lines, 8 programs, \$6,438/week, \$334,800/year — which is the 30% food-cost target expressed as a purchasing budget.
- Poultry: 24 birds Tuesday, 28 Friday = 52 a week = \$582.40, covering a 96-half forecast with an eight-half cushion that becomes Sunday's hash.
- One week's usage: begin \$6,120 + net purchases \$6,510 − transfers \$55 net − ending \$5,930 = \$6,645 used** on \$21,860 of food sales = 30.4% (the shortcuts said 30.7% and 29.8%).
- Turnover: average inventory \$6,025, 1.10 turns a week ≈ 57 a year, 6.4 days on hand.
- A walk-in at 46°F instead of 38°F: \$84 a week, \$4,368 a year, 0.4 points of food cost — against a \$400 repair that pays back in under five weeks.
- Opening inventory: \$9,200 food + \$11,500 beverage = \$20,700.
Key terms
purveyor / vendor · broadline distributor · specialty distributor · product specification (spec) · par level · order guide · receiving · invoice reconciliation · credit memo · prime-vendor agreement · FIFO · physical inventory · the usage formula · inventory turnover · shrinkage
What you should be able to do Monday morning
Stand at the door for the first delivery with a scale, a thermometer, the spec sheet, and the order guide; open every case; and get a signed credit for the first thing that is wrong. Then print the count sheet in shelf order, walk the walk-in with it, and write down the temperature.