How to Track Food Cost Percentage for Restaurants (Step-by-Step)

Tracking food cost percentage means turning sales, recipes, and purchases into one number (and a trend) that shows whether your kitchen is using food as planned, or leaking margin through waste, theft, or bad yields.

Indian full-service restaurants, QSRs, and cloud kitchens often know their menu prices but not their true food cost after aggregator discounts, portion drift, and volatile commodity prices. A repeatable process makes food and beverage finance planning and inventory control speak the same language. For the definitional view of metrics and dashboards, see what is food cost analytics. For how menu mix ties to margin, see menu engineering analytics.

Step 1: Set up ingredient tracking

You need every raw material tied to a stable code, unit of measure, and last purchase price.

• Item master: SKUs for vegetables, proteins, dry stores, and packaging with base UOM (kg, litre, piece) and conversion rules (e.g. 1 crate of tomatoes to kg)
• Vendor mapping: which supplier, price, and tax treatment apply; refresh prices on each GRN or invoice, not once a quarter
• Stores structure: central kitchen, outlet, or cloud-kitchen bin locations so issues and counts roll up to the right P&L

POS and aggregator hygiene: map every selling button to a recipe (or a modifier tree). If "extra cheese" has no ingredient link, theoretical food cost will always lie.

Step 2: Calculate recipe cost

Recipe cost is the sum of ingredient quantities × cost per unit, including normal prep loss (trim, evaporation) where you track it.

Recipe cost per portion = Σ (ingredient qty in recipe × unit cost) + allocated semi-finished items

Roll up to menu price:

Food cost % (item level) = Recipe cost ÷ Net selling price (after discounts) × 100

Net selling price matters in India because platforms and bank offers change what hits your account. Use the price the kitchen should plan against (net revenue per item), not only the printed MRP.

Sub-recipes: gravies, sauces, and batters should be costed as batches, then allocated by gram or portion to final dishes so one gravy price change updates every dependent item.

Step 3: Track actual versus theoretical (AvT)

Theoretical food cost is what you should have used, implied by what you sold:

Theoretical usage = Σ (quantity sold of each item × recipe quantity per sale)
Theoretical food cost value = theoretical usage × standard cost

Actual food cost comes from:

• Issues from stores to kitchen (or auto-deduct from POS-integrated inventory)
• Purchases minus closing stock over the period (period food cost for finance)

Compare:

AvT variance = Actual usage (or cost) − Theoretical usage (or cost)

Slice variance by category (proteins, dairy, beverages), outlet, shift, or station so you fix the right process, not only the total percent.

Step 4: Identify and act on variance

Common drivers to rule in or out:

• Yield and prep: trimming norms not followed; batch sizes wrong
• Portion control: unweighed freehand service; inconsistent scoop sizes
• Pilferage and unrecorded comps: missing void or staff meal discipline
• Pricing and discounts: items sold below the costed floor during campaigns
• Recipe drift: kitchen substitutes without updating the system

Set review cadence: daily flash for QSR (high volume), weekly for full service, with a full period close aligned to finance. Tie actions to owners (head chef, outlet manager, procurement).

Step 5: Automate with POS and inventory data

Automation is repeating the same joins without human copy-paste: POS checks → recipe explosion → purchase and stock movements → dashboard.

Minimum viable automation:

• Live sales feed from POS or aggregator exports
• Recipe engine that updates when cost changes
• Stock ledger from issues, transfers, and counts (even if counts are weekly)

Natural language checks: operators should ask "What was North Indian gravy variance last week?" or "Which outlet is above 34% food cost?" and get an answer from the same data finance uses.

How FireAI helps

FireAI connects POS, recipes, purchasing, and stock movement so food cost percentage and AvT variance stay tied to operations, not a monthly Excel close. You can ask conversational questions about cost by outlet, category, or campaign, and share one view with kitchen and finance. For a broader platform view, see best BI tools for food and beverage in India and hospitality analytics in India.

Deeper operational detail lives in F&B finance use cases and F&B inventory use cases.