Chapter 1 — Key Takeaways
The core claims
-
Ninety percent of restaurants do not fail in year one. The published research puts it near 26%, with roughly 60% gone within three years, where "failure" includes ownership change. The correction matters: most casualties are businesses that worked and then bled slowly.
-
The losses cluster in years two and three, after the opening capital is spent, the novelty traffic normalizes, and the first lease escalation arrives. That is a bleeding pattern, and bleeding is measurable, diagnosable, and preventable.
-
Restaurants keep three to six cents on the dollar. Every consequence in this book follows from that. At a four-point margin, a three-point cost error is not a variance — it is the business.
-
Prime cost is the number that predicts survival. It is most of your controllable cost, its two halves trade off against each other so neither can be watched alone, and it has a benchmark that means something.
-
Compute prime cost weekly, not monthly. Monthly reporting with a three-week lag means learning about a problem seven weeks after it began. Weekly means eight days.
-
Four mechanisms do the actual killing: undercapitalization, cost drift, labor, and cash timing. Every technique in the following thirty-nine chapters is a countermeasure to one of them.
-
Press, cuisine, price point, and culinary talent do not predict survival. Measurement, somebody who understands the numbers, a respected cash cushion, schedule control, retention, and honesty about slow shifts do.
-
The second visit is where the business lives. A first visit is expensive to acquire; a second costs nothing and contributes the same margin. That is why hospitality is a revenue model, not a soft skill.
The formulas
$$\text{Prime cost \%} = \frac{\text{COGS} + \text{Total labor}}{\text{Total sales}}$$
$$\text{Average check (PPA)} = \frac{\text{Sales}}{\text{Covers}}$$
$$\text{Annual revenue} \approx \text{seats} \times \text{turns} \times \text{average check} \times \text{operating days}$$
$$\text{Food cost} = \frac{\text{beginning inventory} + \text{purchases} - \text{ending inventory}}{\text{food sales}}$$
The benchmarks (rules of thumb — ranges, not laws)
| Prime cost | Reading |
|---|---|
| ≤ 55% | strong — likely QSR/fast casual, or exceptional full service |
| 55–60% | healthy full service; the target |
| 60–65% | workable but tight; no cushion |
| 65–70% | distressed; profit mostly gone |
| > 70% | the business is consuming itself |
| Category | Typical full-service range |
|---|---|
| COGS | 28–33% |
| Labor | 30–36% |
| Occupancy | 6–10% |
| Other operating | 12–18% |
| G&A | 2–5% |
| Operating profit | 3–10% |
Key terms
prime cost · cover · average check / PPA · full service · fast casual · quick service · independent · chain · undercapitalization · restaurant failure rate
What you should be able to do Monday morning
Take any restaurant's weekly sales, cost of goods sold, and payroll, compute its prime cost in under a minute, and say whether that business has a cushion or not. Then ask the one question that reveals whether the operator actually knows their food cost: "How do you know? Did you count?"