A higher food-cost percentage tells you that something changed.
It does not tell you what changed.
The cause could be a legitimate commodity increase. It could also be a contract price that did not reach the invoice, a different pack size, an unapproved substitution, an added charge, a missed credit, or a location buying outside the approved order guide.
Those causes can produce the same symptom on a profit and loss statement. They require different fixes.
That distinction matters because restaurants have very little room for unexplained cost movement. The National Restaurant Association estimates that food and labor each consume about 33 cents of every restaurant sales dollar, while the typical restaurant retains roughly 5 cents in pretax profit. The Association also estimates that total expenses for an average restaurant increased 36% between 2019 and 2026. (National Restaurant Association)
In a business with a 5% pretax margin, a recurring purchasing variance does not need to look dramatic to matter.
But a variance is not automatically an overcharge, and an identified opportunity is not automatically savings. The first job is to prove what happened.
Market inflation is context, not invoice proof
Food inflation is real. It is also uneven.
The USDA Economic Research Service forecast food-away-from-home prices to increase 3.5% during 2026. Inside the same forecast, beef and veal prices were expected to increase 9.4%, pork 1.2%, poultry 1.0%, dairy 0.1%, and eggs to decline 29.4%. (USDA Economic Research Service, September 2026 Food Price Outlook)
Those are broad retail price measures and forecasts. They are not benchmarks for a restaurant's contracted distributor price. They do show why the statement “the market went up” is incomplete. Different categories can move in different directions and at very different speeds.
The same is true over shorter periods. From July to August 2026, USDA reported that retail beef and veal prices declined 0.6% while egg prices increased 3.5%. At the producer level, USDA reported that wholesale beef prices declined 3.0% from July to August while remaining 3.1% above the prior year.
All three statements can be true at once.
The market can be higher than last year and lower than last month. A restaurant can also have a contract, deviation, rebate, freight term, or product specification that makes its expected price different from any national average.
National data can help an operator ask a better question. It cannot answer whether one invoice line is correct.
Seven reasons an invoice price changes
1. The underlying market moved
Some agreements use a market index, cost-plus formula, or periodic price reset. In those arrangements, movement may be valid. The review should still identify the contractual basis, effective date, calculation method, and any caps or notice requirements.
“The market increased” is a starting explanation. The agreement should show how that market movement reaches the price on the invoice.
2. The contracted price did not reach the invoice
A negotiated price can exist in an agreement, bid file, or distributor pricing system without appearing correctly on every invoice.
Possible causes include an expired price record, incorrect effective date, wrong customer account, location setup error, missed deviation, or a product that was not mapped to the agreement.
The comparison must connect the agreement to the same item, location, invoice date, unit, and pack.
3. The pack or unit changed
A case price can fall while the unit cost rises. A case price can rise while the underlying unit cost remains unchanged.
Consider two cases:
- Case A costs $48 and contains twelve 32-ounce units. That equals $0.125 per ounce.
- Case B costs $46 and contains ten 32-ounce units. That equals $0.14375 per ounce.
The second case appears $2 cheaper. On a unit basis, it is 15% more expensive.
This is why case price alone is a weak comparison. The product, pack count, unit size, and unit of measure must match. GS1's global identification standard requires a new Global Trade Item Number when declared net content changes and requires a new GTIN at the case level when case quantity changes. That standard reinforces a practical purchasing rule: a different quantity is a different comparison. (GS1 GTIN Management Standard)
4. A substitution changed the economics
The original item may have been unavailable, but the replacement may carry a different brand, specification, yield, pack, or price.
The substitution can be operationally necessary and still deserve review. The questions are:
- Who approved it?
- Was it temporary?
- Does it meet the same specification?
- What is the usable cost after trim, yield, or portion differences?
- Did the approved item return to availability?
A replacement that remains on the order guide after the shortage ends can turn a temporary exception into a permanent cost.
5. The location purchased outside the approved guide
Multi-unit groups often negotiate centrally while ordering locally. That creates a gap between the price the company negotiated and the product a location actually bought.
An off-guide purchase may be reasonable because of availability, guest needs, or a local operating constraint. It may also be a duplicate product, habit, or workaround that was never reviewed.
The purpose of the analysis is to make the difference visible and assign an owner. It should not assume every local exception is wrong.
6. A fee, rebate, deviation, or credit is missing
Invoice price and net cost are not always the same number.
Freight, fuel, split-case charges, early-payment discounts, manufacturer rebates, distributor deviations, and later credits can change the final economics. A purchase may look correct at the invoice line and still be wrong after the full agreement is considered. The reverse can also happen: a high invoice price may be offset by a valid rebate paid later.
The review needs to state whether it is measuring gross invoice price or final net cost. Mixing the two creates false exceptions.
7. The comparison itself is wrong
Restaurant data often contains duplicate descriptions, vendor-specific item numbers, changing units of measure, credits posted in a later period, and purchases split across accounts.
Before escalating an exception, test the data. Confirm that the comparison uses the same product, pack, location, date, purchase unit, and contractual basis.
A spreadsheet can calculate a difference perfectly and still compare the wrong things.
The evidence needed to verify a price increase
A useful invoice-price review connects six records:
| Record | What it establishes |
|---|---|
| Invoice line | What the restaurant was billed, when, and at which location |
| Item master or order guide | What product and pack the location was expected to buy |
| Contract, bid, or pricing agreement | The agreed price or pricing method |
| Product identifier and specification | Whether the compared products are actually the same |
| Rebate, deviation, and credit records | Whether invoice price differs from final net cost |
| Receiving and purchase history | Whether the issue occurred once or became a pattern |
The investigation should be repeatable. Another person should be able to follow the item, location, date, agreement, and calculation and reach the same conclusion.
What the financial result should say
Consider an illustrative example involving twelve restaurant locations.
The group buys an average of 40 cases per location each week. An initial review finds that the invoice price is $2 per case above the price shown in the agreement.
The annualized calculation is:
12 locations × 40 cases × $2 × 52 weeks = $49,920
That $49,920 is an identified opportunity. It is not yet a verified result.
Further review finds that the same correction applies at eight locations, where actual volume averages 35 cases per week and the valid difference is $1.75 per case.
8 locations × 35 cases × $1.75 × 52 weeks = $25,480
That is the value of the implemented change, assuming the correction is put into the pricing system and ordering process.
Eight weeks of post-change purchasing data then show average volume of 32 cases per week at those eight locations, with the $1.75 difference removed.
8 locations × 32 cases × $1.75 × 52 weeks = $23,296
That is the verified annualized impact based on observed post-change purchasing. It is still an annualized figure, not cash already received.
Separately, the distributor issues $6,720 in credits for prior invoices. That is cash recovered.
The four numbers should remain separate:
- Identified opportunity: $49,920
- Implemented change: $25,480
- Verified annualized impact: $23,296
- Cash recovered: $6,720
Combining them would produce a larger headline. It would also make the result harder to trust.
Why this matters across multiple locations
The National Restaurant Association's 2024 operating data, drawn from more than 900 operators, found that food and nonalcoholic beverage costs represented a median 31.0% of sales for full-service restaurants with annual sales of at least $2 million. The median was 33.7% for those below $2 million. Median pretax income was 4.3% and 1.1%, respectively. The Association cautions that these figures are management references, not standards for any individual restaurant. (National Restaurant Association Restaurant Operations Data Abstract analysis)
That warning is useful. A company-wide food-cost percentage should direct attention, but it should not decide the diagnosis.
Two locations can report the same food-cost percentage for different reasons. One may have higher purchase prices. Another may have waste, mix, portion, transfer, or inventory issues. One location's vendor variance may come from the market. Another's may come from an off-guide product.
The percentage identifies where to look. The purchasing records prove what to fix.
A weekly invoice-variance rhythm
Multi-unit groups do not need to investigate every price change with the same intensity. They need a consistent way to prioritize exceptions.
Each week:
- Compare invoiced price with the expected price or pricing method.
- Normalize the comparison for product, pack, unit, location, and date.
- Rank exceptions by dollars, recurrence, and number of locations affected.
- Assign each material exception to an owner.
- Classify the cause as market, agreement, substitution, guide compliance, fee, rebate, credit, or data.
- Record the corrective action and effective date.
- Verify the result in later invoices and purchasing data.
This turns invoice review from a monthly explanation into a purchasing control.
Frequently asked questions
What is restaurant invoice price variance?
Restaurant invoice price variance is the difference between the price billed for a product and the price the restaurant expected to pay under its agreement, pricing formula, order guide, or prior approved purchase. A valid review compares the same item, pack, unit, location, and effective date.
Does a higher invoice price mean the restaurant was overcharged?
No. The increase may reflect a valid market adjustment, substitution, pack change, fee, or contractual reset. The agreement and item-level records are needed to determine the cause.
Can CPI or commodity reports prove a vendor increase is correct?
No. Public indexes provide useful market context, but they do not incorporate a restaurant's specific product, contract, pack, freight terms, rebates, location, or effective date.
What files are needed for a restaurant invoice audit?
The core files are invoice detail, item master or order guide, vendor agreement or bid, product specifications, rebate and deviation records, credit history, and location-level purchasing data.
How often should a restaurant group review invoice variance?
A weekly review allows teams to catch recurring exceptions before they repeat across another purchasing cycle. The review can prioritize material exceptions instead of researching every small movement.
Find the variance. Prove the cause. Verify the result.
BusStop Hospitality helps multi-unit restaurant groups find, prove, fix, and verify margin leakage across vendors, purchasing, pricing, rebates, credits, contracts, order guides, invoices, and location buying behavior.
The Procurement and Margin Recovery Diagnostic begins with the operator's own purchasing records. It maps the spend, identifies material exceptions, tests the evidence, prioritizes corrective action, and establishes how the financial result will be verified.
Find the leak. Prove it. Fix it. Verify the result.
---
Need to verify what is changing across vendor pricing and purchasing?
The Procurement and Margin Recovery Diagnostic maps purchasing data, identifies material exceptions, tests the evidence, prioritizes corrective action, and establishes how the financial result will be verified.
Learn about the diagnostic