Chapter 31 — Key Takeaways

The core claims

1. The restaurant P&L has its own shape for a reason. Group costs by who controls them and on what timescale, then subtotal where a manager can act. That produces cost of sales and labor at the top, prime cost as a subtotal, controllable separated from non-controllable, and occupancy on its own line.

2. Prime cost must be a subtotal on a page somebody reads. A statement that reports cost of sales and labor correctly but never adds them together will let an operator watch food cost drift four points while labor drifts four more — and never see the eight.

3. Compute COGS properly, with the inventory adjustment. Beginning + purchases − ending, then adjust for transfers between kitchen and bar, employee meals at cost, and comped product. The adjusted figure is the one comparable to a recipe cost card.

4. The "purchases ÷ sales" shortcut converges over a year and lies over a week — and it rewards the destructive behavior of managing the timing of purchases rather than the usage of product.

5. Comps, voids, and discounts distort every percentage above them. Keep comps in contra-revenue. A void does not reduce sales and is an audit item, which makes it the most dangerous of the three because it appears nowhere on the statement.

6. Sales tax is a liability, not revenue. It sits in the account making the balance look healthier than it is until it is remitted. Bellwether's is 7%, illustrative.

7. The weekly flash report is the artifact that proves you are managing rather than reporting. Produced within one business day, deliberately fast and approximate, it does not wait for the bookkeeper and does not accrue. It is a complement to the monthly statement, not a replacement.

8. The flash target must ramp. Chapter 9 established Q1 runs at ~66.6% prime and weeks 14–52 must average 58.0%. A flat 60% target would make the report lie for six months — reading green in a quarter that is bleeding and red in one that is fine.

9. Compare like periods. A month is not a comparable unit; five-Friday months beat four-Friday months for reasons that have nothing to do with operations. Use 4-4-5 or 13 periods, or compare the same week year over year.

10. Headcount, scheduled positions, and FTE are three different quantities and none substitutes for another. Publish FTE or hours in a financial document; publish headcount in a hiring plan.

11. EBITDA excludes the principal portion of debt service, which is the omission that gets operators into trouble — principal is cash out and never appears as an expense.

The reconciliations this chapter published

The arithmetic

$$\text{Food cost of sales} = \text{BI} + \text{purchases} - \text{EI} - \text{transfers out} + \text{transfers in} - \text{employee meals} - \text{comped product}$$

$$\text{Prime cost \%} = \frac{\text{cost of sales} + \text{total labor}}{\text{net sales}}$$

$$\text{Controllable income} = \text{revenue} - \text{cost of sales} - \text{labor} - \text{other controllable}$$

Standing conventions: payroll taxes 9.25% of wages · workers' comp 2.90% · total wage burden 20.5% · blended hourly \$14.70.

Key terms

profit and loss (P&L) statement · restaurant chart of accounts · cost of goods sold (COGS) · controllable cost · non-controllable cost · EBITDA · accrual accounting · cash accounting · 4-4-5 calendar · 13-period calendar · weekly flash report · sales tax as a liability · comp · void · discount

What you should be able to do Monday morning

Produce a one-page flash report for last week — sales, covers, cost of sales, labor, prime cost against a ramped target, comps and voids, and the cash reconciliation — inside two hours, without calling the bookkeeper. Then answer the only question that matters: is prime cost inside target this week, and if not, which of the two halves moved?