Food Cost Percentage: The Formula, a Worked Example and What to Do When It Moves

What food cost percentage is, and why it matters in dollars

Food cost percentage is the cost of the food you used in a period divided by your food sales for the same period, expressed as a percentage.

That sounds like accounting. In practice it is the biggest cost line you can change this week. A restaurant doing $104,000 a month in food sales that runs at 34% instead of 31% spends an extra $3,120 a month on ingredients, which is $37,440 a year. For most independents that is the difference between a year that pays the owner and a year that does not. Rent is fixed by a lease and wages move slowly, but food cost reacts within days to a supplier invoice, a heavier hand on the grill or a walk-in nobody rotates.

The percentage matters more than the dollar figure because it is comparable. $32,000 of food cost means nothing on its own. 31% of food sales can be compared with last month, with your target, and with what restaurants like yours usually run.

The food cost formula

There are two versions of the calculation. One tells you what happened in the whole kitchen over a period. The other tells you what a single dish should cost. You need both, and the gap between them is where the money goes missing.

Period food cost (actual)

Beginning inventory is the value of food on hand on the first day of the period, which is the same number as last period's ending count. Purchases are every food invoice received during the period, including the cash run to the store. Ending inventory is what is left on the shelves, in the walk-in and in the freezer when you close the period. Food sales are net of sales tax, tips, and ideally beverages, which belong on their own line.

Plate food cost (per dish)

Plate cost is the sum of every ingredient that leaves the kitchen on that plate, at the price you actually pay, adjusted for trim and cooking loss. The menu price is what the guest pays before tax.

Actual vs theoretical food cost

Theoretical (or ideal) food cost is what your food cost would have been if every plate had been made exactly to its recipe and nothing had been wasted, spoiled, stolen or given away. You get it by multiplying each dish's plate cost by the number sold, adding it up and dividing by food sales. Actual food cost is the inventory calculation above: what really left the building.

If your menu mix says 28.4% and the inventory count says 31.1%, the 2.7 points are real money: on $104,000 of food sales that is about $2,800 a month. A variance under 1 point is a well run kitchen. Two to three points is common and worth chasing. Above four points, something specific is wrong and it is usually findable.

Worked example: costing a cheeseburger with fries

Here is a 6 oz cheeseburger with fries on a casual menu, costed line by line. The prices are illustrative 2026 US wholesale prices from broadline distributor invoices; yours will differ, so use your own last invoice for each line.

1. List every component that goes on the plate, including the ones that feel too small to count: sauce, pickles, fryer oil, the ketchup ramekin.
2. Take the price per purchase unit from the latest invoice ($ per lb, per case, per each).
3. Correct for yield. A pound of iceberg at $1.60 gives about 12 oz of usable leaf after the core and outer leaves, so the usable price is $1.60 ÷ 0.75 = $2.13 per lb.
4. Convert to the portion. 6 oz of ground beef at $5.20 per lb is 6 ÷ 16 × $5.20 = $1.95.
5. Add up the lines to get plate cost, then divide by the menu price.

$3.91 ÷ $14.50 × 100 = 26.97%. The patty alone is half the plate. That is typical: in most dishes one or two ingredients carry 60 to 70% of the cost, and that is where your attention goes. If this burger goes out for takeout, add the clamshell, the fry bag and the napkins; $0.40 of packaging moves it to $4.31 and 29.72%. You can run your own dishes through the food cost calculator, which converts per lb and per each prices to the portion for you.

The monthly period calculation

Now the whole kitchen. A 70 seat casual restaurant closes September with these figures:

$9,400 + $31,800 − $8,900 = $32,300, and $32,300 ÷ $104,000 × 100 = 31.06%. Two details decide whether this number is worth anything. First, the counts must be on the same basis every time: same day, same time (before the first delivery), same unit prices. Second, purchases must belong to the period. An invoice for a delivery on September 30 that you enter in October makes September look 1 point better and October 1 point worse, and you will spend a week chasing a problem that does not exist.

Food cost benchmarks by concept

These ranges are orientation from practice, not a survey. They describe where well run independents of each type commonly land. Your own trend matters more than any benchmark.

The differences come down to four things: how much of the plate is protein, how many ingredients the menu carries (every extra SKU is another thing to spoil), how much labor goes into the food (a scratch kitchen can run a lower food cost because it pays for it in wages), and pricing power. That last point is why food cost % on its own is not the goal. A fine dining plate at 36% on a $48 entrée leaves $30.72 of gross profit; a pizza at 22% on $18 leaves $14.04. Always look at food cost next to labor, which is what prime cost does.

Scenarios: what to do when the number moves

Scenario 1: ground beef goes up 18%

Your distributor moves 80/20 ground beef from $5.20 to $6.14 per lb. On the 6 oz patty that is $1.95 to $2.30, so the burger plate goes from $3.91 to $4.26 and from 26.97% to 29.38% at $14.50. You sell 1,400 burgers a month, so the increase costs $490 a month, or $5,880 a year, from one menu item. Three ways to respond:

• Raise the price. To get back to 26.97%, divide the new plate cost by the old percentage: $4.26 ÷ 0.2697 = $15.80. On a menu, $15.75 lands at 27.05%, close enough. The increase is $1.25, which is 8.6%; guests notice that on a burger, so consider doing part of it now and part next quarter.
• Change the spec. A 5.5 oz patty at the new price costs $2.11 instead of $2.30, taking the plate to $4.07 and 28.07%. That recovers more than half the increase ($0.19 of $0.35) without a price change. It works only if the burger still looks right on the bun.
• Re-engineer the plate. Swap the $0.55 brioche for a $0.38 potato bun and serve 5 oz of fries instead of 6 ($0.42 instead of $0.51). Plate cost becomes $4.00, food cost 27.59%, and the price stays at $14.50.

In practice most operators combine two of these: a smaller price move plus one spec change. What you should not do is nothing and hope the price comes back, because by the time it does you have lost three months of margin.

Scenario 2: food cost jumps 5 points month over month

Last month 30%, this month 35%, on $90,000 of food sales. That is $4,500 of food cost you did not expect. Before you call a staff meeting, check the causes in this order, because the first ones are the most common and the fastest to rule out:

1. Count errors. Was the ending inventory counted in full, including the dry storage and the freezer at the back? Was a case of short rib priced per lb when it should be per case? A single missed $1,500 freezer shelf is 1.7 points on $90,000.
2. Invoices in the wrong period. A late invoice from last month entered this month inflates this month and flatters the last one. Average the two months: if they come out near your normal level, you have a timing problem, not a cost problem.
3. Supplier prices. Compare the unit price of your top 10 items with last month. If beef, chicken, dairy and oil went up 6% and they make up 60% of spend, that alone explains about 1 point.
4. Portioning. Weigh ten plates of your top three sellers during a busy service. A 6 oz portion that is really 7 oz is a 17% overuse on that ingredient.
5. Waste. Look at what was thrown out: prep made for a busy weekend that never came, produce that turned, returned plates. If there is no waste log, that is the finding.
6. Theft and unrecorded comps. Food leaving the back door, meals given to friends and not rung in, manager comps without a reason code. Compare comps and voids in the POS with the previous month.
7. Sales mix shift. If a 24% dish sold less and a 38% special sold well, food cost rises with nobody doing anything wrong. Theoretical food cost for the month tells you how much of the jump is mix.
8. Discounts and staff meals. A 2 for 1 promotion or happy hour food menu lowers food sales without lowering food used. Staff meals that are not rung in at cost count as food cost with no matching sale.

Scenario 3: raising a menu price to restore the target

A chicken entrée costs $4.20 to plate and sells for $14, which is 30%, your target. Over six months poultry and oil prices take the plate to $4.83, a 15% increase, and the dish is now at 34.5%. The price that restores 30% is plate cost divided by target:

On a menu that is $15.95, which is 30.28%. The gross profit after food on each plate goes from $9.17 to $11.12. That means you could sell 17.5% fewer of this dish and still make the same dollars after food cost, which is a useful number to have when someone on the team worries that the increase will cost covers.

The bigger effect is on break-even. Suppose the whole menu drifts from 30% to 33% food cost and your other variable costs (hourly labor, card fees, supplies) take 29% of sales. Contribution margin is 100% − 33% − 29% = 38%, and with $38,000 of monthly fixed costs you need $100,000 of sales to break even. Restore food cost to 30% and the margin is 41%, so break-even falls to $92,683. The same restaurant needs $7,317 less revenue every month before it makes a dollar. Put your own figures into the break-even calculator to see how far a food cost change moves yours.

Tracking food cost over time

A monthly food cost number arrives too late to act on. By the time you count on the 30th and see 35%, the loss has already happened for four weeks. Monthly is the minimum for the P&L; weekly is what lets you manage it.

• Weekly: count the 15 to 25 items that make up about 80% of your food spend (proteins, cheese, oil, the expensive produce) on the same day before the first delivery. Enter invoices as they arrive. A partial count is not perfect, but the trend is reliable and a 2 point move shows up in week one instead of week four.
• Monthly: full count of everything, reconcile purchases against statements, calculate actual food cost, compare with theoretical, and close the period.
• Quarterly: re-cost every recipe at current invoice prices. Plate costs that are a year old are fiction in a market where beef can move 18% in a season.

The other reason a single number is not a system: 31% tells you nothing without context. 31% after three months at 28% is an alarm. 31% after three months at 34% is a recovery. What you need is the number next to last month, next to the same month last year, next to theoretical, and next to labor, so you can see prime cost. That is exactly what a spreadsheet makes painful, because every month means copying tabs and every copied formula is a chance to break the whole file.

This is the job master.restaurant is built for: you enter revenue and expenses by category (with recurring entries for rent and salaries), record the food inventory count, and the dashboard shows food cost and prime cost for the month, the trend against previous months, a break-even worksheet, a month-end projection and plate costing for your recipes. It does not connect to a POS; numbers are entered by hand. It is a one-time payment rather than a subscription.

10 ways to lower food cost percentage

Each of these is concrete and measurable. The dollar figures use the restaurant above, with $104,000 of monthly food sales, where 1 point of food cost is $1,040 a month.

1. Portion with scales and scoops, not by eye. A burger patty that drifts from 6 oz to 6.5 oz costs $0.16 more per burger at $5.20 per lb; on 1,400 burgers that is about $228 a month from one item.
2. Re-cost your top 10 dishes every quarter. They usually make up half of food sales; if one moved from 27% to 33% without a price change, you want to know in March, not December.
3. Get competing bids on your top 10 purchased items. These are typically 60 to 70% of spend. A 4% better price on $20,000 of monthly purchases is $800 a month.
4. Keep a waste log on the line and in prep. Most kitchens that start one find $40 to $80 a day of waste; cutting $60 a day is $1,800 a month, or 1.73 points.
5. Check deliveries against the invoice. Weigh proteins and count cases before the driver leaves. Short weight of 3% on $6,000 a month of protein is $180 a month you pay for and never receive.
6. Ring in every staff meal and comp. $1,200 a month of unrecorded staff meals is 1.15 points of food cost with no sale behind it. Recorded, it becomes a known benefit instead of a mystery.
7. Engineer the menu mix. If 200 orders a month move from an $18 dish at 38% ($6.84 cost) to an $18 dish at 26% ($4.68 cost), you keep $432 more a month at the same revenue. Placement, server recommendations and specials do this.
8. Set par levels and rotate stock. Carrying $12,000 of inventory where $9,000 would do means more spoilage and $3,000 of cash sitting in the walk-in. First in, first out, with dates on every container.
9. Cross-use ingredients. Brisket trim into chili, day-old bread into croutons, herb stems into stock. A menu of 90 SKUs where 70 would do has 20 more ways to spoil.
10. Raise prices in small, regular steps. A 3% increase across the menu takes a 31.06% food cost to 30.16% with nothing else changing. Small steps twice a year cause less guest resistance than one 8% jump after costs have run away.

To check what a change does to a single dish before you print a new menu, run the new portion or price through the food cost calculator and compare the result with your target.