Prime Cost Formula for Restaurants: A Weekly Diagnostic Guide

What belongs in restaurant prime cost?

Use this core calculation:

**Prime cost = cost of goods sold (COGS) + total labour cost**

**Prime cost percentage = prime cost ÷ corresponding sales × 100**

“Corresponding” matters. Monday-to-Sunday sales should not be compared with a food figure that includes the following Monday’s delivery or a payroll figure covering a different fortnight. A polished percentage built from mismatched periods is still unreliable.

Define each input once and keep the definition stable:

Do not quietly change classifications to improve the result. If paper goods sit in COGS this week, for example, moving them elsewhere next week creates apparent improvement without changing the operation.

• **COGS:** food and beverage consumed during the period, classified according to the restaurant’s accounting policy. Where inventory is counted, a practical calculation is opening inventory plus purchases minus closing inventory, adjusted for documented transfers or other classifications used by the business.
• **Total labour cost:** the labour categories the restaurant has chosen to manage in this measure. This may include hourly wages, salaries, employer payroll costs and benefits where applicable. State what is included rather than assuming everyone uses the same definition.
• **Sales:** the sales base used by the restaurant’s accounts. Decide how discounts, refunds, service charges and taxes are treated, document the rule, and apply it consistently to comparisons.

The CLEAR weekly diagnostic

Use **CLEAR** to turn a formula into a controlled review rather than a reflex to cut:

1. **C — Cut off the period.** Write down the exact start and end times. Align sales, inventory counts, receiving, invoices and labour to that cutoff. 2. **L — Lock the definitions.** List what counts as COGS, labour and sales. Note any change from the prior period instead of burying it in the comparison. 3. **E — Execute the calculation.** Calculate food cost, labour cost and prime cost in currency first, then calculate each as a percentage of the same sales figure. 4. **A — Attribute the movement.** Bridge the change into food, labour and sales-denominator effects. Check timing and classification before naming an operational cause. 5. **R — Respond with one bounded action.** Assign a specific check or change, an owner and a review date. Protect service, food safety, quality and team workload while testing the response.

This order is deliberate. It prevents a receiving cutoff error from becoming a purchasing initiative, or a low-sales week from becoming an automatic reduction in essential station coverage.

Reusable weekly prime-cost reconciliation worksheet

Copy this table into a spreadsheet or weekly operating note. Retain the underlying source reference for every figure so another manager can reproduce it.

| Worksheet field | Current week | Previous comparable week | Validation note | |---|---:|---:|---| | Period start and end | | | Same cutoff for every input? | | Sales | | | Treatment of tax, discounts and refunds unchanged? | | Opening inventory | | | Count time and valuation method consistent? | | Purchases received | | | Missing or late invoices accrued or flagged? | | Closing inventory | | | Same locations and units counted? | | Adjustments/transfers | | | Document direction and reason | | Food-and-beverage COGS | | | Opening + purchases − closing, with documented adjustments | | Hourly labour | | | Same work dates, not merely payment date? | | Salaried labour allocation | | | Allocation method unchanged? | | Employer costs and benefits included | | | Same categories each week? | | Total labour cost | | | Sum of included labour categories | | Prime cost | | | COGS + total labour | | Food cost percentage | | | COGS ÷ sales × 100 | | Labour cost percentage | | | labour ÷ sales × 100 | | Prime cost percentage | | | prime cost ÷ sales × 100 | | Change from comparison period | | | Show currency and percentage-point movement | | Unusual events | | | Closure, event, count correction or other context | | Data-quality status | | | Pass, provisional or restate | | Diagnosed component | | | Food, labour, sales base, mixed or unresolved | | One follow-up action | | | A check or controlled intervention, not a vague goal | | Owner and review date | | | Named accountability and due date |

“Previous comparable week” should mean operationally comparable, not simply the row above. If opening hours, event business or count methods differ materially, retain the week in the record but label the comparison limitation.

Example: a Monday-to-Sunday full-service restaurant

This is an illustrative calculation, not a benchmark or a recommended target. Assume the restaurant closes its reporting week on Sunday night and validates that all figures cover those same seven days.

| Current-week input | Amount | |---|---:| | Sales | $48,000 | | Opening food-and-beverage inventory | $7,600 | | Purchases received | $13,200 | | Closing food-and-beverage inventory | $6,800 | | Hourly labour | $10,900 | | Salaried labour allocated to the week | $3,800 | | Included employer payroll costs and benefits | $1,300 |

First calculate COGS:

**$7,600 + $13,200 − $6,800 = $14,000**

Then calculate total labour:

**$10,900 + $3,800 + $1,300 = $16,000**

Prime cost in currency is:

**$14,000 + $16,000 = $30,000**

Prime cost percentage is:

**$30,000 ÷ $48,000 × 100 = 62.5%**

For comparison, suppose the prior comparable week had $50,000 sales, $14,000 COGS and $15,000 labour. Its prime cost was $29,000, or 58.0% of sales. The current week therefore moved by **$1,000** and **4.5 percentage points**.

The component bridge is more informative than the combined movement:

The next step is not “cut 4.5 points”. The reviewer should first validate whether the extra labour belongs to the week and whether the sales decline reflects comparable trading days. If the data passes, a bounded follow-up could be: “Operations manager to compare scheduled and worked hours by daypart, identify the source of the $1,000 increase, and report before the next schedule is approved.” That action investigates a component without presuming that understaffing is the answer.

• Food remained $14,000, but moved from 28.0% to about 29.2% because sales fell.
• Labour increased by $1,000 and moved from 30.0% to about 33.3%.
• The combined percentage rose because labour dollars increased while the sales denominator decreased.

Decision table: what to investigate next

| What the worksheet shows | First check | Suitable next action | Avoid | |---|---|---|---| | Data-quality check fails | Cutoffs, missing invoices, inventory units, work dates and classifications | Correct or mark the week provisional; recalculate | Treating accounting noise as operational failure | | Food dollars rise; labour is stable | Purchase prices, quantities, inventory movement, transfers, waste records, portions and sales mix | Select one evidenced food-cost cause for a controlled check | Assuming all variance is waste or raising every price | | Labour dollars rise; food is stable | Worked versus scheduled hours, overtime, prep, training, coverage and sales by daypart | Investigate one shift, role or process while setting service gates | Cutting coverage evenly across all shifts | | Both components rise | Shared causes such as a launch, event, longer opening hours or weak sales; then inspect each component | Prioritise the largest validated driver and sequence responses | Launching unrelated food and labour cuts at once | | Dollars are stable; percentage rises | Sales definition, closures, discounts, volume and trading hours | Diagnose the sales denominator before changing cost controls | Calling stable spending a cost overrun without context | | One unusual week drives the change | Inventory correction, payroll allocation, closure, event or delayed invoice | Annotate it and compare with another valid period | Rewriting the operating plan from one distorted observation | | Movement persists with clean data | Component trend and the restaurant’s own plan | Form a hypothesis, choose one measure and set a review date | Using a generic industry threshold as proof of the cause |

If food remains the validated source of movement, RestaurantMargin’s [30-day food-cost plan](https://restaurantmargin.com/reduce-food-cost) provides a related actual-versus-theoretical reconciliation workflow. Keep the prime-cost worksheet as the combined control; use the deeper food analysis only after the split points there.

Implementation steps for a dependable weekly review

1. Set one reporting clock.

Choose the weekly opening and closing timestamps. Schedule the inventory count and data extraction around that clock. Record exceptions rather than silently stretching the period.

2. Publish a one-page definition sheet.

List included COGS and labour accounts, the sales basis, inventory valuation method and treatment of transfers. Give the sheet a version date. When a definition must change, calculate or label the break in comparability.

3. Reconcile inputs before calculating.

Confirm inventory locations, invoice receipt dates, labour work dates and sales adjustments. A “provisional” status is more useful than false precision when an input is missing.

4. Record currency and percentages together.

Currency shows the amount to explain. Percentages show the relationship to sales. Retaining both helps distinguish increased spending from a smaller denominator.

5. Build the component bridge.

Compare current food dollars and percentage with the chosen prior period, then do the same for labour. Note the sales change separately. Do not let the combined percentage hide offsetting movement—for example, food improvement alongside labour deterioration.

6. Add operational context.

Document changed opening hours, menu launches, events, closures, inventory corrections and payroll adjustments. Context should explain what to test; it should not be used to dismiss an unfavourable result without checking it.

7. Assign one primary follow-up.

Write the action as an observable task: what will be checked or changed, by whom, by when, and which measure and guardrails will be reviewed. Where a change could affect service, safety, availability, quality or workload, include those guardrails explicitly.

8. Close the loop next week.

At the next review, state whether the task was completed and what the comparable measure did. If several things changed at once, acknowledge that attribution is weak. Keep the record even when the hypothesis was wrong; it prevents repeated guesswork.

Weekly review checklist

Before approving an intervention, confirm:

• [ ] Sales, inventory, purchases and labour cover the same period.
• [ ] COGS, labour and sales definitions match the comparison period.
• [ ] Missing invoices, transfers and count corrections are resolved or flagged.
• [ ] Prime cost is shown in currency and as a percentage.
• [ ] Food, labour and sales-denominator movement are separated.
• [ ] The comparison period is genuinely comparable, or its differences are stated.
• [ ] Unusual events are documented.
• [ ] The proposed action addresses a validated component.
• [ ] Service, safety, quality, availability and workload guardrails are defined where relevant.
• [ ] The action has an owner and review date.

Limitations and tradeoffs

Prime cost is a compact operating measure, not a complete account of restaurant performance.

Do not use this weekly method as a substitute for the formal accounts, payroll review, food-safety controls or professional accounting advice appropriate to the business.

• **It is not profit.** It excludes whichever operating, occupancy, financing, tax and other expenses sit outside the restaurant’s COGS and labour definitions. A prime-cost result cannot establish net profitability by itself.
• **It is not cash flow.** Inventory consumption, invoice timing, payroll payment dates and capital or financing movements can make cash behave differently from the weekly cost calculation.
• **There is no universal target in this guide.** Concepts differ in service model, product mix, pricing, labour structure, opening hours and overhead. Use the restaurant’s operating plan and comparable history; do not treat the illustrative example as a threshold.
• **The total can hide causes.** Two restaurants—or two weeks—can have the same prime-cost percentage with very different food and labour patterns. Component analysis is mandatory before choosing a response.
• **Weekly data can be noisy.** Count errors, events, closures, accrued invoices and payroll allocations can distort one period. Faster review is valuable only when cutoff discipline is strong.
• **Classification choices affect comparability.** Including or excluding a labour-related cost changes the number. Consistency and disclosure matter more than pretending there is one universal chart of accounts.
• **Cost action has operational tradeoffs.** Reduced coverage can damage service or increase workload; altered purchasing or portions can affect availability, safety, consistency or guest value. Prime cost identifies where to investigate, not which sacrifice to make.
• **Aggregate analysis can miss local problems.** A weekly total may not reveal a weak daypart, station, category or menu item. Move to a more detailed analysis only after the aggregate data is validated.