Ignore the average restaurant margin. It cannot help you.
Short answer: your margin is decided by four numbers. Prime cost as a percent of sales, contribution margin per menu item, waste and comps, and labor scheduled against actual covers instead of habit. Fix those in that order. And if you're only going to do one thing this month, raise prices before you cut costs, because a price move drops straight to the bottom line and a cost cut usually costs you something else.
You've read that the average restaurant runs 3 to 5 percent net. That figure gets quoted everywhere and it's useless to you. It's an average across a Denver steakhouse doing 400 covers on a Saturday and a breakfast spot in Berthoud that closes at two. It doesn't tell you whether your Tuesday lunch shift makes money. It doesn't tell you that your best-selling item might be your worst-earning item.
I owned a bakery. I know exactly what it feels like to have a great sales month and a checking account that says otherwise. The money was never missing. It was leaving in four places, every day, in amounts small enough that I didn't feel any single one of them.
Prime cost is the one ratio you run every week
Prime cost is cost of goods sold plus total labor, including your payroll taxes and your own wage if you're working the line. Divide it by sales. That single percentage tells you more than a P&L your bookkeeper hands you six weeks late.
The common industry rule of thumb: full-service kitchens aim for prime cost around 65 percent of sales or lower, and quick-service or counter-service tends to target closer to 60. Call it a rule of thumb, not gospel. What matters is your own trend line. If your prime cost was 62 percent in May and it's 71 percent in August, something moved, and you want to know that in seven days, not in ninety.
Here's the honest math on why this is urgent. Say you do $60,000 a month. Every point of prime cost is $600. Nine points is $5,400 a month, which is $64,800 a year walking out the back door while you're congratulating yourself on a busy patio season.
Run it weekly. Sales from the POS, purchases from your invoice stack, labor from payroll. Fifteen minutes, Monday morning, coffee in hand. That's it.
Cost your menu item by item, not by category
Most owners know their overall food cost percentage. Very few know which items earn and which items freeload. That's where the money is hiding.
At the bakery, I costed things out by item, and it changed what I made. Two products would sit next to each other in the case at similar prices, and one of them earned real money while the other one barely paid for the butter, the labor, and the case space it was taking up. Same shelf, same customer, wildly different outcome for me.
Do this: pull your top 20 sellers. For each one, write the plate cost (every ingredient, including the garnish and the oil it fried in) and the menu price. Subtract. That difference is contribution margin in dollars, and dollars is what you deposit, not percentages.
Now sort your 20 items into four buckets:
- Sells a lot, earns a lot. Protect it. Put it where eyes land first.
- Sells a lot, earns little. This is the trap. Reprice it, reengineer the portion, or change what comes with it.
- Sells little, earns a lot. Train the staff to recommend it. Move it up the menu.
- Sells little, earns little. Kill it. It's costing you prep time, freezer space, and a SKU you have to keep in stock.
A hypothetical to make it real: an item priced at $16 with a $5.60 plate cost gives you $10.40 per plate. An item priced at $14 with a $6.30 plate cost gives you $7.70. Sell 200 of the second one instead of the first and you left $540 on the table that week. Nobody notices $540. Fifty-two weeks of it is a used truck.
Schedule labor against covers, not against habit
Labor is the number that gets away from you quietly, because the schedule is a copy of last week's schedule, which was a copy of the week before. Nobody decided anything. It just kept happening.
In Northern Colorado this is seasonal in ways that will lie to you. Fort Collins gets CSU move-in weekend and home game Saturdays and everybody staffs up. Greeley has Stampede. Windsor and Johnstown have summer patios that make August look like your business is fundamentally better than it is. Then January arrives and the same schedule is still running.
Patio months are the worst offender because volume hides bad scheduling. You're up 30 percent on sales and up 35 percent on labor and it still feels like a great month. It isn't. That's a losing trade.
The fix is boring and it works. Pull covers by day-part for the last eight weeks out of your POS. Build the schedule against that pattern, not against who wants hours. Set a labor target per shift in dollars, not bodies, and let the manager cut early when the rush doesn't show. I scheduled a rush that didn't come plenty of times at the bakery. Everybody stood around getting paid, the product I'd prepped for it went in the trash at close, and I ate it twice.
Waste and comps are the leak nobody logs
Waste doesn't appear on any statement. It shows up as food cost being higher than your recipes say it should be, and then you go argue with your produce vendor about a price that wasn't the problem.
Get a clipboard by the trash can. Every item thrown out, written down, for two weeks. Nothing fancy. What that list gives you is a prep par sheet based on what you actually sell rather than what somebody guessed in 2023. At the bakery, throwing product away at close was the single most demoralizing part of the day, and it was also the most fixable, because the answer was making less, not selling more.
Comps and voids get the same treatment. Pull the comp report by employee and by reason. Not to run anybody off. To find out whether you're comping a dish because the recipe is inconsistent, or whether one server is buying friends drinks. Both are real. Both are your money.
If waste plus comps is running 2 to 3 percent of sales and you get it to 1, on that $60,000 month that's $600 to $1,200 back with no new customers and no new menu.
Raise prices without losing your regulars
Here's the part owners resist and shouldn't. A price increase is the highest-return move available to you, because it costs nothing to implement and every dollar of it is margin.
Run the numbers on a hypothetical. You do $60,000 a month at a 3 percent net, so you keep $1,800. Raise prices 4 percent across the board, your costs don't move, and you now keep about $4,200 on $62,400. You more than doubled your take-home without touching a single vendor contract.
Would you lose customers? Some, maybe. Run the break-even: if a 4 percent price increase costs you 4 percent of your covers, you're still ahead, because the covers you lost were carrying food and labor cost with them. Do the math for your own numbers before you talk yourself out of it.
How to do it so it lands:
- Don't raise everything the same amount. Raise the low-margin high-sellers and leave the signature item alone. Regulars watch one or two prices, not thirty.
- Change the menu at the same time. New layout, new descriptions, two new items. A new menu with new prices reads as a new menu. A sticker over the old price reads as a price hike.
- Move in odd amounts. $14 to $15.25 is invisible in a way $14 to $16 is not.
- Tell your staff why, before a guest asks. A server who apologizes for the price teaches the guest to be annoyed.
I'm not trying to blow smoke. Raising prices is uncomfortable the first week and then nobody mentions it again.
The one-page scorecard you'll actually keep up
How do you eat an elephant? One bite at a time. Don't build a dashboard. Build one page you fill out Monday morning in fifteen minutes:
- Sales, last week
- Cost of goods (invoice total), last week
- Total labor including taxes, last week
- Prime cost as a percent of sales, and last week's number next to it
- Covers by day
- Waste and comp dollars
Six lines. Print it, write on it, stick it in a binder. Four weeks of that page will tell you more about your restaurant than a year of monthly P&Ls, because you'll see the pattern while you can still do something about it.
Then pick one number a month. This month, prime cost. Next month, the bottom five menu items. The month after, the schedule. Trying to fix all four at once is how owners fix none of them.
If you want a second set of eyes on which of the four is actually costing you the most, take twenty minutes with the free Business Checkup or grab a 30-minute call and bring last month's numbers. If the fix is something you can do yourself in an afternoon, I'll tell you that and you can go do it. If it's bleeding badly enough that it needs hands on it this week, that's what Triage is for. Flammable, or on fire.
Your margin lives in four numbers: prime cost weekly, contribution margin per item, waste and comps, and labor scheduled against real covers. And when you're deciding where to start, a 4 percent price increase almost always beats a month of cost cutting.
Questions I get about this
What is a good restaurant profit margin?
The commonly quoted average is 3 to 5 percent net, but that average is across every format in the country and won't tell you anything useful about your place. Track prime cost instead: cost of goods plus all labor, divided by sales. Rules of thumb put full service around 65 percent or lower and counter service nearer 60. Your own trend line week over week matters far more than any benchmark.
Should I cut food costs or raise prices first?
Raise prices first. A price increase costs nothing to implement and every dollar of it lands in margin, while cost cutting usually trades quality or hours for savings. On a $60,000 month, a 4 percent increase adds roughly $2,400 in gross profit if your costs hold steady. Cut costs after, once you know which items and shifts are actually losing money.
How often should I calculate prime cost?
Weekly. Monthly P&L statements arrive too late to change anything, and by the time you see a bad month it's already spent. Fifteen minutes on a Monday with your POS sales, invoice stack and payroll gets you the number while you can still act on it.
What's the fastest way to find waste in a restaurant?
Put a clipboard by the trash can and log everything thrown out for two weeks, then log comps and voids by employee and reason. Those two lists rewrite your prep pars and usually explain the gap between what your recipes say food cost should be and what it actually is. It costs nothing but attention.
How do seasonal swings in Northern Colorado affect margins?
Patio months and event weekends in Fort Collins, Greeley and Windsor inflate sales enough to hide bad labor scheduling, then January arrives with the same schedule still running. Build your schedule off eight weeks of actual covers by day-part and set a labor dollar target per shift so managers can cut early when the rush doesn't come.
Bring me last month's sales, invoices and payroll, and in thirty minutes I'll tell you which of the four numbers is eating your money.
Thirty minutes, no pitch. If I can't help, I'll tell you who can.
Book a 30-minute call or start with the free Business Checkup →