A coffee shop runs on thousands of small transactions and two big percentages — cost of goods and labor. This guide sets up the daily and weekly rhythm that keeps both percentages true.
Hi, I'm Max — I run Vickery Finance, and this page is one room in the workshop. Everything here walks the same path: from shoebox-and-guesswork to a clean close you can defend, one honest number at a time. Work through it with your own figures in hand — that's when a guide turns into a result. And when you're ready to do the close instead of just reading about it, start with the Free Month-End Close Checklist: it's page one of the 18-step close my templates run, printable and yours to keep. Know your numbers.
Cost of goods percentage — beans, milk, syrups, pastries, cups and lids against sales — tells you whether drink prices still cover ingredient costs that have been quietly rising. Labor percentage tells you whether the schedule matches the dayparts it serves. Prime cost, the two combined, is the shop's health check, because it captures the costs management actually controls week to week.
Everything here is organized to produce those numbers on a rhythm a small shop can sustain: sales proven daily against the point-of-sale system, invoices entered as they arrive, and a short weekly count on the high-value ingredients (beans and milk) that drive the goods percentage.
Fifteen minutes a day beats a lost weekend at month-end.
Spilled drinks, remakes, expired milk, and staff drinks are real costs of goods that never became sales. Log them in a simple daily waste note (count and reason). Without it, the goods percentage absorbs waste invisibly and pricing decisions get made on the fiction that every ounce was sold. With it, waste becomes a manageable line — many shops find 2–4% of goods cost hiding there.
A neighborhood shop does $31,500 in sales for the month. Invoices and counts give cost of goods of $10,080; payroll gives total labor of $9,450 including taxes and benefits; rent is $3,200:
| Measure | Math | Result |
|---|---|---|
| Sales | POS month total | $31,500 |
| Cost of goods % | $10,080 ÷ $31,500 | 32% |
| Labor % | $9,450 ÷ $31,500 | 30% |
| Prime cost | 32% + 30% | 62% |
| Rent | $3,200 ÷ $31,500 | 10% |
| Left for other costs and profit | 100% − 62% − 10% | 28% |
One day's POS report says $1,120 in gross sales. The proof agrees it to settlements before anyone asks questions:
The Coffee Shop Profit and Inventory Tracker is built around this rhythm — daily sales, goods and labor percentages, prime cost, and bean-and-milk inventory in Excel or Google Sheets. The restaurant bookkeeping guide runs the same system for a full-service kitchen.
Companion to this guide: the Coffee Shop Expense Tracker — beans, milk, waste, and fees built into a true cost per cup, with a worked cafe week.
Many operators watch prime cost in the high-50s to mid-60s percent of sales, but the right target depends on rent, price point, and whether food is a sideline or a pillar. Treat benchmarks as conversation starters — your own month-over-month trend is the number that manages the shop.
A gift card sale is a liability (you owe drinks), not revenue. Revenue is recognized when the card is redeemed. Keep a simple card liability tally and reconcile it monthly — breakage and stale balances follow your state's rules, so flag them for your accountant rather than sweeping them into sales.
No. The rhythm above runs fine in a spreadsheet — that is the premise of every template on this site. Software earns its keep when volume, multiple locations, or payroll complexity outgrows the weekly routine.
A coffee shop's year is only as true as its last inventory count. On December 31, after close, I count everything: beans by the bag, milk, syrups, cups, lids, and pastries. Say the count values stock at $3,240, while the books — opening inventory plus purchases minus the month-by-month estimates — were carrying $3,890. That $650 difference is a COGS true-up: it means waste, over-portioning, or shrinkage was higher than my weekly numbers assumed, and December absorbs the correction so the new year starts from a count I can defend. A shop that skips the count does not avoid the cost; it just spreads the lie across twelve months of margins.
Gift cards are the second December trap, and the holidays make it a big one. If I sold $3,850 in gift cards in December and $1,120 of the shop's outstanding cards were redeemed that month, none of the $3,850 is December revenue — it is a liability that converts to revenue only as cards are redeemed, and the outstanding balance (including what carried in from earlier months) must sit on the balance sheet at year-end, net of any breakage my state law and policy actually allow. I also reconcile the sales tax / TPT collected on taxable sales to what was remitted, pull payroll records together for W-2s in January, total any 1099-NEC amounts for unincorporated vendors I paid $600 or more, and list equipment for the depreciation schedule — the $6,500 espresso machine repair that replaced a boiler is a capital conversation, while the $180 gasket kit is a repair, and December is when I sort one from the other with invoices in hand.
My closing checklist: full physical inventory counted and COGS trued up; gift card liability reconciled card by card to the POS report; sales tax collected tied to filings; payroll and 1099 totals ready for January; and prime cost — COGS plus labor — computed for the full year as the benchmark next year's menu prices must beat.
Staying open two hours later sounds like found money until the labor line speaks. I tested my own shop's idea with a month of till data. The proposed window — 5 p.m. to 7 p.m., 26 days a month — would need two staff on the floor. Projected sales for those hours, based on the last hour currently traded, are $68 an hour: $68 times 52 hours is $3,536 a month. Costs follow automatically. COGS at my shop's 32% is $1,132. Labor is two people at $17 an hour for 52 hours: $1,768. Payroll taxes and burden at 12% of wages add $212, and lights, water, and machine wear for the extra hours run about $60. Total added cost: $3,172. The extension earns $364 a month — if the $68-an-hour projection holds.
That margin of safety is thin: at $60 an hour instead of $68, the same hours lose money, and no projection survives contact with a quiet Tuesday. So the decision the books support is a limited trial, not a commitment — extend Friday and Saturday only (8 evenings, 16 staffed hours per person), where my POS already shows the strongest late trade, measure actual sales per labor hour for six weeks, and make the full extension only if the trial clears $75 per labor hour with waste counted. Hours are a labor purchase. I buy them in the smallest quantity that produces a trustworthy number.
1 — Your first win (10 minutes): the Free Month-End Close Checklist — page one of the 18-step close, printable, free.
2 — Do it with the workbook: find your trade's template — every workbook runs this same clean-close ritual, with example data and a guided first step inside.
3 — Keep the rhythm: one practical numbers email a week — Know Your Numbers, free to subscribe.
The way back: start any visit at the Start Here hub, or email a human — [email protected]. I answer.
Know your numbers. — Max
Some links below are affiliate links. If you buy through them, Vickery Finance may earn a small commission at no extra cost to you.
Adams Guest Check Pads, 2-Part Carbonless, 10-Pack — for restaurants and food trucks: order pads that leave a copy for the kitchen and the books.
Adams Invoice Book, 2-Part Carbonless, 50 Sets — for invoicing on paper in the field: write it once, leave a copy, keep a copy.
Brother Genuine TN-730 Standard-Yield Toner — the standard-yield factory toner for Brother TN-730 machines.
Epson WorkForce ES-50 Portable Document Scanner — for scanning receipts and documents one page at a time, anywhere.
More books and gear live on the Resources page.