Reducing Food Waste in Restaurants: A Practical Playbook
Beyond the headlines — concrete tactics for inventory counting, par levels, and prep sheets that move waste from 7-9% of sales down to under 3%.
By NASS Editorial · · 10 min read · Inventory, Sustainability
Restaurants in 2026 waste, on average, between 5% and 10% of their food purchases. For a restaurant doing $1.2M a year on a 30% food cost, that's between $18,000 and $36,000 of pure waste annually — money that goes straight from the bank account to the bin.
This piece is a practical playbook for cutting that number in half within a single quarter, written for operators who do not have the time or budget for a full consulting engagement.
Stop guessing your par levels
The number one cause of avoidable waste is over-ordering. The number one cause of over-ordering is gut-feel par levels. "Order enough chicken thighs to get us through the weekend" is not a system.
Start with three weeks of point-of-sale data, broken down by day of the week. For each menu item, calculate average and 90th-percentile sales by weekday. Multiply by your recipe yields. That gives you a realistic par for each ingredient by day, with a buffer for unusually busy days.
Update those numbers quarterly. They will drift as your menu and marketing change.
FIFO is only useful if you can see it
"First in, first out" is a principle every kitchen knows and almost no kitchen actually executes well. The reason is purely physical: when a fresh delivery arrives during prep, the cooks shove it on top of yesterday's stock because there is no time to rotate. Three days later, yesterday's product is at the bottom, hidden, and forgotten.
The fix is mechanical, not motivational: - Pre-cut date stickers on every container, every day. - Walk-in shelves labelled by section, not by ingredient — "next to use" sections at eye level. - A daily 5-minute walk-through by a manager, with permission to throw out anything past date on the spot.
The cost is 15 minutes a day. The savings are usually 2-4% of food cost.
Prep sheets that match yesterday's sales, not last week's
A common pattern: the prep cook makes a fresh batch of demi-glace every morning because "we usually use a litre". Some days you use 0.4 L. The rest goes in the bin on day 3.
Modern POS systems can export sales of every menu item, by item, by day. Use them to build a prep sheet that scales prep volumes to *yesterday's actual sales of dishes that use this ingredient*, not a static "make 1 L". The math is simple; the discipline of using the data daily is the hard part.
The 90/10 rule for menu engineering
When a dish sells fewer than 10 covers a week, the ingredients it requires are almost guaranteed to spoil. Run a quarterly menu audit:
- Identify the bottom 10% of dishes by sales volume.
- For each, ask whether the ingredients are also used by other dishes. If yes, keep. If no, cut.
- Replace cut items with seasonal specials that use ingredients already in the walk-in.
This single exercise typically removes 3-5 items from the menu, eliminates 6-10 unique ingredients from the stock list, and reduces ordering complexity in a way that compounds.
Train staff to over-portion, then measure
Counterintuitive but consistently effective: do not punish cooks for over-portioning during a service. Punish unmeasured portioning. Issue scales at every station and require a weekly spot-check audit. Track variance against the spec.
When cooks learn that their portions are being weighed (not penalised, weighed), variance drops by 30-40% within two weeks, and food cost drops with it.
Trim, peel and stock production
The cheapest protein on most menus is "stock". A kitchen that throws away bones, chicken carcasses, vegetable trim, and herb stems is throwing away a future menu item.
A weekly four-hour stock production session, scheduled into the prep calendar, typically reclaims $200-$400 of edible value from what would otherwise be waste. Bones go to brown stock; vegetable trim to vegetable stock; herb stems to oils and vinegars; bread to crostini and breadcrumbs.
The trick is scheduling. If stock production is not on a written prep sheet with a name and a time, it does not happen.
Inventory by exception, not by ritual
Counting every shelf every week is exhausting and unsustainable. Restaurants that do it well count the *expensive 20%* of items twice a week (proteins, seafood, premium produce) and the rest monthly. Combine with weekly theoretical-vs-actual variance reports from the POS, and you catch problems early without spending a full Sunday in the walk-in.
A measurement that actually changes behaviour
Most operators measure food cost as a percentage of sales. That is the right number for the P&L, but it is a terrible operational signal because it moves for too many reasons (menu mix, price changes, weather).
Instead, measure *waste log entries per cover* — the count of items thrown out, divided by total covers, by day. Track it for two weeks, post the number on the kitchen wall, and watch it fall. Cooks respond to numbers they can see and influence directly.
The savings, summarised
A kitchen that does these things consistently reduces total waste from 7-9% of sales down to 2-4%. For a typical independent restaurant, that's $20,000-$40,000 a year falling straight to the bottom line.
None of these tactics require new software, new hardware, or new staff. They require attention, consistency, and the willingness to look at uncomfortable data every week. That is harder than buying a tool, which is exactly why so few restaurants do it — and why the ones that do enjoy a quiet, structural advantage.