Eight ways a restaurant P&L lies to you
Most restaurant P&Ls are accurate and still misleading. These are the eight distortions worth checking before you trust yours.
None of these are fraud. Every one of them shows up in books kept honestly by people doing their best. They're distortions of structure rather than of fact, and they're worth knowing because each one changes a decision.
1. The owner works for free
The most common distortion in independent restaurants, and the largest. If you're expediting five nights a week and taking distributions instead of a salary, your labor line is understated by whatever it would cost to replace you — call it $65,000 to $90,000 for a working GM.
The profit at the bottom isn't profit. It's your wages, relabeled. This matters enormously the day you try to sell, hire a manager, or take a step back, because all three convert an invisible cost into a real one, and the business that looked marginally profitable becomes plainly unprofitable overnight.
2. Repairs near zero
Repairs and maintenance under half a point of sales is not cost discipline. It's deferral. The walk-in compressor doesn't care that you had a tight year, and the bill arrives as capital expense at a moment you don't choose. A suspiciously clean R&M line is a liability that hasn't been booked yet.
3. Delivery commissions hidden in G&A
Third-party delivery fees frequently land in general and administrative, where they sit next to software subscriptions and insurance and disappear. Commission at 15–30% of order value is not an administrative expense — it's a cost of sale on that channel, and at any real volume it sets your margin rather than your menu does.
Give it its own line. Operators who do this routinely discover they're running an unprofitable second restaurant inside the profitable one, priced for dine-in and sold at a commission.
4. Utilities counted as occupancy
Occupancy benchmarks of 5–9% of sales exclude utilities. Combine them and a healthy lease reads as an expensive one. This is worth checking before any conversation about renegotiating or relocating, because it's a distortion that prompts expensive decisions.
5. Employee benefits that are too small
Payroll taxes, workers' compensation and benefits generally run 12% or more of wages, and restaurant workers' comp rates are not low. If your benefits line is materially under that, some of those costs are sitting somewhere else, and your labor cost is understated by the difference.
6. Food cost calculated from invoices
Purchases divided by sales is not food cost, though it's often labeled that way. It ignores inventory movement entirely: a week where you build stock looks terrible and a week where you draw it down looks excellent, and neither reflects anything the kitchen did. Real cost of goods needs a physical count. Without one you're managing noise.
7. Non-alcoholic beverages in the beverage line
Standard practice puts soda, coffee and juice in food sales and food costs. Moving them to beverage inflates your pour cost and starts an investigation into a bar that's running fine. Small distortion, disproportionate amount of wasted attention.
8. Percentages against the wrong base
Food cost belongs against food sales, beverage cost against beverage sales. Run either against total sales and your mix starts masquerading as performance — a strong beverage quarter makes the kitchen look improved when nothing changed. The error is silent and self-flattering, which is the worst combination.
Why these persist
Almost every one of these comes from the same root: the statement was produced by someone who is good at accounting and has never run a shift. A generic chart of accounts is perfectly correct as bookkeeping and blind to the things that decide whether a restaurant survives.
Your bookkeeper isn't doing anything wrong by putting delivery fees in general and administrative — that's a defensible classification. It's only wrong in the sense that it hides a number you need to see weekly. The fix is almost never a new bookkeeper; it's a fifteen-minute conversation about the chart of accounts, and it's usually welcomed, because most bookkeepers would rather code things the way you'll actually use them.
The one exception is unbooked owner labor, which is a real decision rather than a coding choice. Booking a market-rate salary for yourself makes the statement honest and the reported profit smaller. That trade is worth making before you need the statement to be believed by a buyer, a lender or a partner — all three will normalize it themselves, and it's better to have done it first.
A short checklist
- Is there a management salary on the books, and does it reflect what a replacement would cost?
- Is repairs and maintenance at least 1% of sales?
- Does third-party delivery have its own line?
- Is occupancy free of utilities?
- Are benefits at least 12% of wages?
- Was there a physical inventory count this period?
- Are NA beverages in food?
- Is food cost measured against food sales?
Eight questions. If more than two come back wrong, the statement isn't telling you what you think it is, and every decision you make from it inherits the error.