Understanding Inventory Variance (and What It's Telling You)

Every kitchen runs on a quiet assumption: that the food you bought turns cleanly into the dishes you sold. Inventory variance is the number that tests that assumption. It is the gap between what your recipes say you should have used and what your shelves say you actually used. That gap is rarely zero, and reading it well is one of the highest-leverage habits an independent operator can build.

This post walks through what inventory variance is, why the two numbers almost never match, what a small gap versus a large gap is telling you, and how to turn the whole thing into a weekly routine instead of a year-end surprise.

What inventory variance actually measures

There are two ways to answer "how much chicken did we use this week?"

Theoretical usage is the recipe answer. Take every dish you sold, multiply by the amount of each ingredient its recipe calls for, and add it all up. If you sold 200 plates that each use 6 oz of chicken, theory says you used 75 lbs.

Actual usage is the shelf answer. Count what you had at the start of the week, add what you received, subtract what you counted at the end. Whatever is missing is what you actually used.

Inventory variance is the difference between the two, usually expressed as a percentage of theoretical usage. If theory says 75 lbs and your count says you burned through 82 lbs, you have a 7-lb variance, roughly 9%. Those seven pounds left the building without becoming a sale.

Why the two numbers rarely match perfectly

A little variance is normal. Trim loss shifts week to week, scales drift, a count happens ten minutes later on Tuesday than it did on Monday. For most items, a small percentage gap is noise, not a signal.

The size of the gap is the first thing to read:

• A small, stable gap (low single digits on most items) is the cost of doing business. Watch it, don't chase it.
• A large gap, or one that keeps growing, means product is leaving unaccounted for. That is money, and it deserves an investigation.

The danger isn't the occasional blip. It's the persistent leak that looks small each week and quietly adds up over a quarter.

The main causes, and how to tell them apart

Variance is a symptom. The work is diagnosing the cause, and the same 8% gap can come from very different problems. The usual suspects:

• Over-portioning. The most common cause and the easiest to miss. If cooks free-pour instead of using a scale or a portion scoop, every plate walks out a little heavy. It shows up as a steady, consistent gap on a specific item, worst on things portioned by hand.
• Waste and spoilage. Product that expired, got dropped, burned, or was prepped and never sold. Tends to spike around over-ordering or slow weeks. A waste log is what separates this cause from the others.
• Theft. Product leaving without a sale, whether it's food out the back door or drinks poured for friends. Often shows on high-value, easy-to-move items (proteins, liquor, cheese) and tends to be persistent rather than random.
• Receiving errors and short deliveries. You were billed for 40 lbs but the case held 36. This one inflates your theoretical stock, so it reads as usage variance later. Weighing key deliveries at the door is the only real check.
• Comps and staff meals not recorded. Food that was genuinely used but never rang up as a sale. The product is gone from the shelf but absent from theoretical usage, so it looks like a loss. It isn't shrinkage; it's a recording gap.
• Recipe or spec drift. Your recipe card says 6 oz but the kitchen has quietly been plating 7. The recipe is now wrong, so theory is wrong, and every plate reads as variance. Re-checking the actual portion against the card catches it.

You tell these apart by pattern. Steady on one item points to portioning or spec drift. Spiky and tied to slow weeks points to waste. Persistent on high-value items points to theft or receiving. That's the detective work variance makes possible.

You can't diagnose what you can't measure

Here's the hard part. Variance is only as trustworthy as the two numbers feeding it.

If your count is sloppy, inconsistent, or done by a different person each week with a different idea of what "one bag" means, your variance is noise and you'll chase ghosts. Count the same way, in the same order, at the same point in the week.

If your recipes are wrong or missing, your theoretical usage is fiction, and a "variance" is really just the distance between reality and a bad spec. Accurate, current recipe cards are the foundation the whole calculation stands on. This is also why variance and your real food cost percentage rise and fall together: both depend on knowing what a plate is actually supposed to contain.

Start with the items that matter

You do not need to track variance on all 300 SKUs. That's how the habit dies.

The 80/20 rule applies cleanly here: a handful of high-cost, high-volume items drive most of your food cost and most of your risk. Proteins, seafood, cheese, oil, alcohol. Get variance tight on your top 10 or 15 items and you've covered the money that actually moves. Paper napkins can wait.

Tight ordering discipline makes this easier, because par levels that survive a Friday night keep your counts in a predictable range instead of swinging wildly week to week.

A simple worked example

Say a hypothetical taco spot tracks its 8 oz carne asada.

• Theoretical usage for the week (tacos sold times 8 oz): 62 lbs
• Starting count + deliveries − ending count: 71 lbs
• Variance: 9 lbs, about 15%

Fifteen percent is far past noise, so they investigate the biggest gap first. The waste log shows only about a pound tossed. Deliveries checked out at the door. But watching the line, cooks are eyeballing portions and plating closer to 9.5 oz. Spec drift plus over-portioning. They retrain to a scale, and next week's recount lands at 4%. The leak is closed, and they move to the next item.

Turn it into a routine

Variance only pays off as a repeatable loop, not a one-time audit:

1. Count your top items consistently, same method every week.
2. Compare actual usage against theoretical.
3. Investigate the biggest gaps first, not every gap.
4. Fix the specific cause you found.
5. Recount next cycle to confirm the fix held.

The recount step is what separates a real fix from a guess. And it's what tells a one-off blip apart from a persistent leak: a blip disappears next week on its own, while a leak keeps showing up until you find and close it.

This week

Pick your three highest-cost items. Confirm the recipe card matches what the kitchen is actually plating, then do one careful count using the same method you'll use every week from now on. Run the comparison, look at the single biggest gap, and ask what pattern it fits. You don't need software or a perfect system to start. You need three items, an honest count, and the willingness to look. The number is already talking. This week is about learning to listen.