Two price baskets. Two different questions.

Read each index as its own price history

The BLS Consumer Price Index separates food at home from food away from home. These baskets describe different consumer purchases. The first is useful context for grocery prices; the second is useful context for prices paid for prepared food away from home. Neither is a direct index of a particular restaurant's procurement costs.

The chart uses the non-seasonally-adjusted US city-average series in our checked extract. It preserves the published index levels and leaves missing observations blank. The reference base is 1982 to 1984 = 100 in the source; here we write that period as 1982 to 1984 to keep the date range clear.

The endpoint change is the later index divided by the earlier index, less one, multiplied by 100. It is a change across the displayed window, not an annual inflation rate. Do not label it year over year simply because the chart contains monthly observations.

Why the level difference is not a markup

The difference between the food-away and food-at-home index levels is not a restaurant markup percentage. Each basket has its own composition and price history. An index describes change relative to a reference base, not the absolute dollar price of a comparable meal.

A restaurant meal includes more than raw ingredients. Preparation, service and the operating environment are part of what the consumer buys. That does not mean the CPI can tell you what any one of those components costs. It is still a consumer-price measure rather than a restaurant cost accounting report.

Compare changes within a series over matched dates. If you want to show relative movement from a common starting point, rebase each series deliberately and label that transformation. Do not quietly alter the source scale or connect missing months as if observations existed.

Use the data to frame a question

The useful operating question is whether your own prices and costs have moved in a way that changes contribution. A broad consumer index can provide context for that review, but your invoice records and recipe quantities determine the ingredient effect on a plate.

A menu containing ingredients with different price histories will not move exactly like the overall basket. The supplier mix, purchase specification and contract also matter. Substituting a retail average for a wholesale invoice without labeling it would make the operating calculation look more precise than it is.

Keep the public series beside your internal record rather than overwriting it. Record when a supplier price changed, which pack it applied to and which recipes use that ingredient. Then test the contribution effect before deciding whether to change a recipe or selling price.

Work a price-pressure example

Assume a dish sells for $15 and has $4.50 of ingredient cost. Its food-cost percentage is 30% and ingredient-level contribution is $10.50. These are example inputs, not a representative restaurant or an observed menu.

Now assume the actual ingredient basket for that dish rises by 8%, while its recipe and selling price remain unchanged. Ingredient cost becomes $4.86. Food cost rises to 32.4% and contribution falls to $10.14. The 8% assumption belongs to this scenario; it is not the BLS index movement shown above.

Preserving contribution dollars and preserving a percentage produce different prices. Neither is automatically the right decision. The business must also consider other costs, expected volume, guest response and the role of the item in the menu.

Separate a price decision from a demand assumption

A spreadsheet can calculate a higher selling price while holding sales volume constant. That is a scenario, not evidence that guests will buy the same quantity. State the volume assumption whenever a model uses it.

Test different expected quantities and inspect total contribution. An item with a higher contribution per sale can produce less total contribution if volume falls enough. Conversely, an item with a small contribution improvement may matter substantially if it is sold frequently. The sales mix determines how individual decisions combine.

Keep discounts and channel fees visible. The printed menu price may differ from the amount retained after a promotion or delivery agreement. A consumer-price index cannot reconcile that difference for your operation.

Missing months and publication dates

The checked extract covers a monthly window ending in September 2026. Some observations are unavailable in the returned source data. They remain null in the stored data and appear as gaps, rather than being replaced with a previous value or a straight-line estimate.

A checked date records when we reviewed the source. It is different from the observation month. The distinction matters when someone returns to the page later: an accurate historical number should not be mistaken for today's supplier quote or the newest release after this review.

The static site does not poll BLS while you use the app. Prefilled examples retain the stated source date until the published dataset is updated. Your working ledger should use your own current purchase records when you make an operational decision.

A repeatable review routine

Choose a consistent date for reviewing supplier changes. Confirm pack sizes and credits before updating prices, then inspect the recipes most affected in dollars. Include finished yield so a preparation change is not hidden inside what looks like price inflation.

Review the selling price, contribution and expected volume together. Keep a copy of the scenario and the final decision. After implementation, compare actual sales and usage with the assumptions rather than attributing every subsequent change to the price adjustment.

Use the menu-price calculator for a chosen food-cost target and the contribution calculator for the remaining ingredient-level dollars. The ingredient catalog provides clearly labeled retail references when you need an external comparison, with each pack basis and source attached.