How to Reduce Food Waste in Your Restaurant: US Guide 2026

The short version. The EPA estimates that 30-40% of the US food supply goes to waste (EPA, Food Loss and Waste). In a restaurant, that waste shows up in your food cost as 3-8% of food spend — silently, consistently, week after week. It breaks down into four families: trim and breakage, spoilage, over-portioning, and shrinkage. Each one has a fix. None of them require starting over.

What restaurant food waste actually costs you

Restaurant food waste is everything you buy that does not end up on a paying plate. Trim you throw out. Product that expires before you use it. Portions that are 15% too heavy because there is no scale within reach. Staff meals you do not log. Dishes you 86 mid-service because ordering was done on instinct.

The USDA estimates that food waste costs US consumers and businesses about $161 billion annually (USDA Economic Research Service). For an independent restaurant running $800K in annual revenue at 32% food cost, even moving from 5% waste to 2% waste is roughly $7,200 back in the pocket.

That is not a sustainability win. That is a payroll decision.

The four families of restaurant food waste

The first mistake is treating all waste as one category called "waste." Each family has its own cause, fix, and owner. Treating them the same means correcting blind.

1. Trim and breakage

Trimming a cut of beef, filleting fish, peeling produce — this generates unavoidable losses. But "unavoidable" does not mean "unmeasurable." A beef strip loin loses 15-20% to trim depending on supplier quality and cook technique. A whole salmon fillet yields about 60% of its raw weight after portioning.

These trim rates belong in your recipe cards. Without a documented yield percentage, you cannot tell if your line cook is trimming efficiently or 10% over.

2. Spoilage and expired product

20% of losses in the 38°F zone are tied to spoilage in an average restaurant operation. That number climbs when ordering is not based on real sales data — you order "just in case" and the product expires before you use it.

An expired product is 100% of its value lost. No recovery. Main driver: over-ordering from your broadliner combined with FEFO not being enforced in the walk-in. New Sysco delivery goes in front of what was already there. The old product expires in the back of the shelf.

Date marking is the operational fix. FDA Food Code § 3-501.17 requires TCS foods to be date-marked within 24 hours and used or discarded within 7 days (day 1 = prep day). A proper date marking system forces your team to see expiration dates — and FEFO happens naturally.

3. Over-portioning

The most insidious waste category. Nobody catches it in real time. The cook plates 8 oz instead of 6 oz because "it looks better on the plate" or because there is no scale within reach of the station. Multiply by 50 covers and 300 service days and the numbers get uncomfortable fast.

I ran this math at La Verrerie in 2016. We had a flank steak selling well. Recipe called for 180g (roughly 6.3 oz). Actual plates weighed 198g on average — 18 grams of variance per plate. Across 40 covers a day, 300 days a year, that was 216 kg of steak going out without being charged. At the purchase price at the time, that was roughly $4,750 walking out the door annually on one dish. No bad faith. No theft. Just no scale within reach.

The fix: a fixed scale at the cold station, a note on the recipe card, one brigade meeting to address it. Two weeks later the variance was back to 3 grams.

4. Shrinkage and unrecorded items

Staff meals not logged, kitchen tasting, comps not tracked, till voids. Not always dishonest — but always a real cost. In a well-run operation, staff meals are costed at purchase price and included in the theoretical food cost. Comps get tracked too. If you do not account for them, they inflate your real food cost without explanation.

How 86'd dishes connect to waste

Every time you 86 a dish mid-service, there is a story behind it. Either you ordered too little (ordering problem) or the dish had poor sell-through early in the week and you over-prepped for a volume that never came (forecasting problem).

Both directions are waste opportunities. Track which dishes you 86 — and when. If you are 86ing the salmon every Thursday, you either ordered light or sold fewer covers than expected Monday through Wednesday and the prep is sitting in the walk-in approaching the 7-day mark.

That 7-day TCS mark under § 3-501.17 is not just a compliance item — it is a waste clock. Every TCS item you prep is on a countdown. Date marking makes that countdown visible. When your team can see it, they act on it.

How to measure your food waste

Two methods. Ideally run both for a test period.

1. The food cost variance method. Calculate your theoretical food cost (what you should have consumed based on sales and recipe cards) and compare it to your real food cost (what you actually consumed via inventory count). The gap is your waste. The breakdown by category tells you where.

2. Waste bin weighing. Weigh your waste bins by category — raw trim, plate waste, expired product — for one test week. Cost what you weigh at purchase price. This is tedious but precise. It tells you exactly where the dumpster is getting expensive.

Waste benchmarks by US operation type

These are 2026 benchmarks cross-referenced with available industry data. Your operation has its own specific mix — a long prix-fixe menu with expensive proteins mechanically pulls the number up.

Common mistakes on food waste

• Counting only visible waste. Over-portioning never hits the bin — it goes through the ticket and shows up in the COGS variance. Invisible if you do not weigh plates.

• Trusting that your line cook knows the right portion. Without a scale at the station and a recipe card they actually consult, even a skilled cook drifts 10-20 grams per plate across a shift.

• Ignoring FEFO. If you stack new deliveries in front of what was already in the walk-in — because it is faster when the Sysco truck shows up — you guarantee spoilage losses on the product pushed to the back.

• Counting only once a month. Once a month you observe. Once a week on high-cost items, you control. Weekly counts on proteins and produce catch losses in the week they happen, not 30 days later.

• Costing staff meals at zero. If staff meals are not costed, they inflate your real food cost without explanation. Include them in your theoretical COGS at purchase price.

Conclusion

Restaurant food waste is not a fate. It breaks down into four families — trim, spoilage, over-portioning, and shrinkage — and each one has a concrete fix.

Three things to hold onto:

1. Over-portioning is the most underestimated lever. Invisible to the eye, impossible to see without weighing, and capable of costing several thousand dollars a year on a single dish. A fixed scale at the station and a recipe card that gets followed are the fix.

2. Date marking is your spoilage early-warning system. FDA Food Code § 3-501.17 is not just compliance paperwork. It is a physical reminder that every TCS item in your walk-in is on a countdown. When your team can see the clock, they manage to it.

3. Measuring waste is already acting on it. A four-week run of food cost variance tracking tells you exactly where to look. Most operators have never run the number. The ones who do are consistently surprised by what they find.

18 grams per plate. $4,750 a year. That is where it starts.

Last updated 2026. Written by Cyril Quesnel, founder of Onrush. 20 years on the line in France, two restaurant turnarounds. Building food safety + food cost tools for US indie restaurants.