Restaurant Bookkeeping Done Right: Prime Cost, Food Cost Percentage, and the Weekly Scorecard

Restaurants fail with profitable-looking monthly P&Ls all the time. The margins are thin enough — a good full-service restaurant keeps 5–10 cents of every dollar — that a problem discovered at month-end has already eaten four weeks of it. Restaurant bookkeeping done right runs on a weekly rhythm, tracks two numbers above all others, and reconciles the POS to the bank every single day.

Key takeaways

• Prime cost — food and beverage cost plus total labor — is the number that decides survival. Healthy full-service restaurants run 55–62% of sales; over 65% for long is a countdown clock.
• Food cost percentage only means something when calculated with inventory counts: beginning inventory + purchases − ending inventory, divided by food sales. Purchases alone will lie to you in both directions.
• Restaurants should close their books weekly (or on 13 four-week periods), not monthly — a calendar month with five Fridays and Saturdays isn't comparable to one with four.
• The POS is the system of record for sales; the daily sales summary must tie to bank deposits, with card settlement timing and third-party delivery fees tracked explicitly.
• Sales tax collected and tips owed to staff are not your money. Book them as liabilities from day one.

The two numbers that matter

Food cost percentage. Take a real count: beginning inventory plus purchases minus ending inventory equals cost of goods used; divide by food sales. Counting weekly is what makes the number honest — a big Friday delivery booked as "purchases" with no count makes a fine week look terrible and the next week look great. Typical ranges: 28–35% for full-service, lower for pizza and some fast-casual concepts, with beverage tracked separately (liquor pour cost 18–24%, wine 30–40%, beer 20–25%). The comparison that matters most isn't the industry number — it's your theoretical food cost (what the POS says you sold, priced at recipe cost) against actual. The gap between theoretical and actual is waste, portioning drift, and theft, and it's found nowhere else.

Prime cost. COGS plus all labor — wages, payroll taxes, benefits, payroll processing. This is the controllable core of the P&L; occupancy is fixed and everything else is small. Full-service concepts should target roughly 55–62% of sales; QSR can run lower on labor-light models. When prime cost runs hot, the weekly cadence tells you which lever moved — a food-cost spike points at purchasing, waste, or theft; a labor spike points at scheduling against a soft sales week.

A chart of accounts built for a restaurant

Sales split by revenue center: food, liquor, beer, wine, N/A beverage, catering, merchandise, and third-party delivery tracked separately (more below). Comps and discounts as contra-revenue — not expenses — so net sales is real.

COGS mirrors the sales split: food purchases by major category (meat, seafood, produce, dairy, dry goods), liquor, beer, wine. That symmetry is what lets you compute pour cost and food cost by center.

Labor split at minimum between front-of-house, back-of-house, and management, with payroll taxes and benefits on their own lines. Owner compensation separate, always.

Then occupancy (rent, CAM, utilities), operating (smallwares, linen, paper, cleaning, repairs), G&A, and marketing. Two liability accounts that must exist from day one: sales tax payable and tips payable. Both are other people's money passing through your bank account; restaurants that treat them as cash flow are the ones that meet the Florida Department of Revenue and the trust fund recovery penalty later.

The daily and weekly rhythm

Daily: post the POS end-of-day summary — gross sales by center, comps, sales tax collected, tips (cash declared and card), payment types — and reconcile expected deposits. Card batches settle in 1–2 days; cash deposits should match the drawer paperwork the same day. Cash variances get logged daily because patterns, not single incidents, are how skimming shows up.

Third-party delivery deserves its own lane. DoorDash, Uber Eats, and Grubhub deposit net of 15–30% commissions, and sales tax handling differs (in Florida, marketplace facilitators generally collect and remit tax on marketplace orders — your books need to reflect who remitted what). Book delivery revenue gross, commissions as expense, and reconcile each platform's statement monthly. Netting it all into "deposits" hides your true sales and overstates margins by center.

Weekly: inventory count, food cost and prime cost calculation, invoice entry (or AP automation via a tool like MarginEdge or xtraCHEF), labor versus schedule review, and a one-page scorecard: sales vs. last year, prime cost, theoretical-vs-actual food cost, labor %, covers or check average. Thirteen four-week periods beat calendar months for all of it — every period has the same day mix, so trends are real.

Monthly: full bank/card reconciliations, loan splits, sales tax return tied to the POS (the Florida DOR audits restaurants by comparing POS z-tapes, purchases, and returns — gaps between POS sales and reported sales are the classic finding), and a P&L with every line as a percentage of sales.

Accrual-ish is good enough

Most independent restaurants file taxes cash-basis, and that's fine. But pure cash-basis management books distort a business that pays vendors on terms — a week where you paid three weeks of invoices looks catastrophic. The practical middle: enter vendor invoices when received (AP-driven COGS), count inventory weekly, and let the tax return be adjusted to cash at year-end. You get accrual-quality operating visibility without audit-level accounting cost.

Official sources

• IRS Publication 334 — Tax Guide for Small Business: https://www.irs.gov/publications/p334
• IRS — Recordkeeping: https://www.irs.gov/businesses/small-businesses-self-employed/recordkeeping
• Florida DOR — Sales and use tax on restaurants (GT-800035): https://floridarevenue.com/forms_library/current/gt800035.pdf

Practitioner note: If your bookkeeper hands you one P&L a month and no prime-cost number, you don't have restaurant bookkeeping — you have tax-return preparation on a delay. The weekly scorecard is the product; the monthly close is just hygiene.

Fairlight builds weekly-scorecard bookkeeping for restaurants across Miami, Hollywood, Fort Lauderdale, and greater South Florida — POS-to-bank tie-outs, prime cost tracking, and sales tax filings included. Contact us or see pricing.