F&BosEvery cloud kitchen founder learns some version of the same lesson, usually in month two or three of trading: the margins that look comfortable in a spreadsheet do not survive the first contact with a live delivery platform. The dish that appears to have a 30% food cost in isolation is actually carrying a 52% effective cost burden once aggregator commission, packaging, and platform-side discounting are factored in. The 20% net margin projection turns into a 6% EBITDA reality.
This is not a spreadsheet error. It is a structural feature of the cloud kitchen model that most business plans account for incorrectly. The food cost formula for a cloud kitchen is not the same formula that applies to a dine-in restaurant. The difference is aggregator commission: a 22-28% fee taken off every order before any other cost is calculated. This single line item transforms the entire cost structure of the business and requires a different approach to food cost calculation, pricing, and margin management.
This article walks through five formulas every cloud kitchen operator needs, builds the complete cost waterfall from gross order value to net profit, explains recipe-level costing in practical terms, and identifies the most common places where food cost calculations go wrong. The benchmarks throughout are calibrated to the Indian cloud kitchen market in 2025-26.
Cloud kitchen cost structure (typical): Platform commission 22-28% | Food cost 28-34% of net revenue | Packaging 3-5% | Labour 8-12% | Rent 5-8% | Net margin 10-25%.
Net revenue retained after platform commission: 70-78% of gross order value.
Prime cost target (food + labour) for cloud kitchens: 40-46% of net revenue. Industry benchmark for restaurants: 60-65% of gross revenue.
Aggregator commissions on delivery orders mean a dish with 30% food cost in isolation carries 45-55% total cost burden relative to gross order value.
The 5 Formulas You Need
From recipe cost to real margin — the complete calculation set
Most food cost guides cover one or two formulas in isolation. A complete cloud kitchen food cost analysis requires five, applied in sequence. Each formula builds on the output of the previous one. Running them in order, using your own numbers, gives you a single source of truth for your unit economics.
Formula 1 Food Cost Percentage
Example: A chicken biryani costs Rs 95 in ingredients. Selling price: Rs 340.
Food Cost % = (95 / 340) x 100 = 27.9% → Target: 25-30% for cloud kitchens.
Formula 2 Effective Food Cost (Platform Orders)
Factor)
Net Revenue Factor = 1 - Platform Commission Rate (e.g. 1 - 0.25 = 0.75 for 25% commission)
Example (same biryani): Rs 95 / (Rs 340 x 0.75) = Rs 95 / Rs 255 = 37.3%
The effective food cost on a platform order is 9.4 points higher than the listed food cost %.
Formula 3 Recipe Cost per Serving
Q Factor = Rs 8-15 per dish (accounts for salt, oil, spices, condiments not in recipe card)
Example: Paneer tikka — paneer 150g x Rs 480/kg = Rs 72 | capsicum 60g x Rs 60/kg = Rs 3.60
Tomato 40g x Rs 40/kg = Rs 1.60 | onion 30g x Rs 25/kg = Rs 0.75 | spices/marinade = Rs 8
Total recipe cost = Rs 85.95 + Q Factor Rs 10 = Rs 95.95 per serving
Formula 4 Prime Cost
Prime Cost % = (Food Cost + Labour Cost) / Net Revenue x 100
Target for cloud kitchens: 40-46% of net revenue (vs 60-65% for dine-in restaurants)
Example: Monthly food cost Rs 1,20,000 + Labour Rs 55,000 = Prime Cost Rs 1,75,000
If net revenue is Rs 4,00,000 → Prime Cost % = 43.75% → Healthy for a cloud kitchen.
Formula 5 Actual vs Theoretical Food Cost Variance
Theoretical = what you would spend if every dish was made exactly to recipe with zero waste.
Actual = what you actually spent (from purchase records and closing inventory count).
Acceptable variance: 2-3 percentage points. Above 4 points = investigate immediately.
Example: Theoretical 28.5%, Actual 33.1% → Variance 4.6% → Rs 18,400/month profit leak at Rs 4L revenue.
The Cost Waterfall
What happens to every rupee from order placement to your pocket
The most useful analytical tool for a cloud kitchen operator is the cost waterfall: a structured breakdown that starts at gross order value — the amount the customer pays — and applies each cost category in sequence until it reaches net profit. The waterfall reveals exactly where margin is being consumed and how much room is left at the bottom.
The example below uses a cloud kitchen doing 40 orders per day at Rs 380 average order value, with 25% platform commission, 30% food cost on net revenue, 4% packaging, 10% labour, 7% rent, and 4% other fixed costs. These are close-to-median figures for a well-managed independent cloud kitchen in a metro city in 2025-26.
| Cost Line | Monthly (Rs) | % of GOV | Notes |
|---|
| Gross Order Value (GOV) | 4,56,000 | 100.0% | 40 orders x Rs |
|---|---|---|---|
| 380 x 30 days | |||
| Less: Platform commission (25%) | (1,14,000) | 25.0% | Swiggy / Zomato / ONDC |
| Net Revenue | 3,42,000 | 75.0% | What hits your bank account |
| Less: Food cost (30% of net rev) | (1,02,600) | 22.5% | Target 28-32% of net revenue |
| Less: Packaging (4% of net rev) | (13,680) | 3.0% | Rs 11-17 per order avg |
| Gross Profit | 2,25,720 | 49.5% | After direct costs |
| Less: Labour (10% of net rev) | (34,200) | 7.5% | 2-3 staff @ Rs 12-18K/month |
| Less: Rent (7% of net rev) | (23,940) | 5.25% | KaaS bay or own premises |
| Less: Utilities & misc (3%) | (10,260) | 2.25% | Gas, electricity, water |
| Less: Marketing (2%) | (6,840) | 1.5% | Platform promo / direct mktg |
| EBITDA | 1,50,480 | 33.0% | Before depreciation & tax |
| Less: Depreciation | (8,500) | 1.9% | Equipment / fit-out over 3 yrs |
| NET PROFIT | 1,41,980 | 31.1% of net rev | ~22.5% of GOV |
|---|
Read the waterfall in two ways: as a % of GOV (gross order value — what the customer paid) and as a % of net revenue (what you actually received). Net margin of 31% of net revenue = 22.5% of GOV. Both figures are correct; they describe different things.
The waterfall above assumes 25% commission. If your blended commission is 28%, your net revenue drops to 72% of GOV and every cost below it (food cost %, labour %) increases as a % of GOV. A 3-point commission difference can move your net margin by 4-6 percentage points.
Building in a direct ordering channel (WhatsApp, ONDC) that takes 10-15% commission vs 25% on Swiggy/Zomato shifts the waterfall significantly. At 20% of orders direct at 12% commission, the blended commission falls from 25% to 23.4% — a Rs 7,500/month saving at this revenue level.
Recipe Costing Step by Step
Calculate the exact cost of every dish before it goes on your menu
Recipe costing is the process of calculating the exact ingredient cost for one serving of a dish. It is the foundation of everything else in this article: you cannot know your food cost percentage, set a defensible price, or track a theoretical vs actual variance unless you have a costed recipe for every item on your menu. The recipe cost is not an approximation — it is calculated to the gram using your actual supplier prices.
Step 1 — Collect Actual Purchase Prices
Use your most recent purchase invoices. Do not use estimated market prices — use the actual price you paid from each supplier in the last week. Ingredient prices in India fluctuate significantly (onions, tomatoes, and cooking oils are seasonally volatile by 30-60%). If you use estimated prices, your recipe cost will be wrong by the same percentage as your price estimation error.
Convert all prices to a per-gram or per-ml unit cost. If you buy paneer at Rs 480 per kg, the per-gram cost is Rs 0.48. If you buy cooking oil at Rs 160 per litre, the per-ml cost is Rs 0.16. Maintaining a unit cost master sheet that you update monthly is the single most impactful food costing habit a cloud kitchen operator can develop.
Step 2 — Weigh Each Ingredient to Recipe Spec
Every ingredient in every dish must have a specified quantity to the gram. Not 'a handful of onions' or 'two tablespoons of ghee' — grams and millilitres, weighed on a kitchen scale. This specification is your theoretical usage per serving. It is the number you apply the unit cost to. It is also the number you train your kitchen team to replicate on every order.
Account for trim and yield loss. If a recipe calls for 100g of raw chicken but you trim and butcher a whole bird with 78% usable yield, the effective cost per 100g of usable chicken is the whole-bird price divided by 0.78. Ignoring trim yield is one of the most common causes of understated recipe cost in Indian restaurant kitchens, where whole vegetables are bought and significant weight is lost in cleaning, peeling, and chopping.
Step 3 — Add the Q Factor
The Q Factor is an allowance for ingredients that are used in the dish but not individually tracked in the recipe card: cooking oil absorbed during frying, salt added to taste, cumin seeds used in tadka, garnishing coriander, lemon wedge. In Indian cooking, where tadka ingredients and spice blends are applied to many dishes, the Q Factor is typically Rs 8-15 per serving for a cloud kitchen. Add it to every recipe cost as a fixed line, not as an approximation.
Worked Example: Chicken Biryani
| Ingredient | Qty (g/ml) | Unit (per kg/L) | Yield Factor | Cost (Rs) |
|---|---|---|---|---|
| Chicken (bone-in) | 350g | Rs 220/kg | 0.72 (trim) | Rs 106.94 |
| Basmati rice | 200g | Rs 120/kg | 1.00 | Rs 24.00 |
| Onion (biryani fried) | 80g raw | Rs 35/kg | 0.75 (trim+fry) | Rs 3.73 |
| Yoghurt (marinade) | 50ml | Rs 80/L | 1.00 | Rs 4.00 |
| Ghee | 15ml | Rs 580/L | 1.00 | Rs 8.70 |
| Saffron + spice mix | 3g blend | Rs 200/100g | 1.00 | Rs 6.00 |
| Mint leaves + garnish | 10g | Rs 60/kg | 0.80 | Rs 0.75 |
| Q Factor (oil, salt, spices, tadka) | — | — | — | Rs 12.00 |
| TOTAL RECIPE COST | Rs 166.12 | |||
| Selling Price | Rs 560 | |||
| Food Cost % | (Rs 166 / Rs 560) x 100 | Target: 25-30% | 29.7% | |
| Effective Food Cost % (25% comm.) | Rs 166 / (Rs 560 x 0.75) | Target: below 40% | 39.6% |
The effective food cost for the biryani above (39.6%) is already at the top of acceptable range. If the platform commission rises to 28%, effective food cost becomes 43.1% — above the danger threshold. This is why the selling price for delivery must be set using effective food cost, not listed food cost.
Trim yield matters. The chicken in this example has a 0.72 yield factor. Operators who skip yield calculation and cost at the purchase price per kg are understating recipe cost by Rs 30-40 on a 350g protein serving.
Pricing from Food Cost
Set your menu prices to protect margins, not guess at them
Once you have recipe costs for all menu items, you have the data to set prices that protect your margins. The standard food cost pricing formula — dividing recipe cost by your target food cost percentage — gives you a minimum acceptable selling price. For a cloud kitchen, this calculation must be performed on effective food cost (post-commission) rather than listed food cost.
The Pricing Formula for Cloud Kitchens
Example: Recipe Cost Rs 95 / 0.35 (target effective food cost 35%) = Rs 271 minimum
selling price
Cross-check: Rs 271 x (1 - 0.25 commission) = Rs 203 net received / Rs 95 cost =
46.8% effective food cost
An alternative framing that many operators find intuitive is the multiplier method: price each menu item at 3.5-4x its recipe cost. At 4x recipe cost with 25% platform commission, effective food cost is 33.3%. At 3.5x recipe cost with the same commission, effective food cost is 38.1%. The 3.5x multiplier should be treated as the floor, not the target. Dishes where the market will bear 4x or more should be priced accordingly.
Pricing Strategy by Menu Tier
| Menu Tier | Recipe Cost | Multiplier | Selling Price | Eff. FC % | Strategy Note |
|---|---|---|---|---|---|
| Loss leader | Rs 40 | 3.0x | Rs 120 | 44.4% | Drives discovery; offset by other items |
| Core dishes | Rs 80-120 | 3.5-4x | Rs 280-480 | 33-38% | Volume items; must hit target FC |
| Premium dishes | Rs 140-200 | 4-5x | Rs 560-1000 | 27-33% | High margin; improves blended FC |
| Sides & add-ons | Rs 15-30 | 4-6x | Rs 60-180 | 20-28% | Best margin items; upsell aggressively |
| Beverages | Rs 10-25 | 5-8x | Rs 50-200 | 15-22% | Highest margin; add to every menu |
A cloud kitchen running 60% of revenue in core dishes at 36% effective FC, 20% in premium dishes at 30% effective FC, 10% in sides at 24% effective FC, and 10% in beverages at 18%
effective FC has a blended effective food cost of: (0.6 x 36) + (0.2 x 30) + (0.1 x 24) + (0.1 x 18) = 21.6 + 6 + 2.4 + 1.8 = 31.8%.
This blended effective food cost of 31.8% is sustainable even if the commission rises to 28%, because effective food cost is already calculated net of commission. The operator has margin headroom.
The operator who prices all dishes at 3.5x recipe cost and runs 80% of revenue in core dishes has a blended effective FC of 38% — workable, but with no headroom for a commission increase or a food cost spike.
Actual vs Theoretical: Finding Your Profit Leak
The variance calculation that tells you where money is being lost
The difference between theoretical food cost and actual food cost is the most important management metric in a cloud kitchen — more useful than revenue growth, more actionable than average order value, and more predictive of long-term profitability than any other single number. The variance represents money that should have been profit but was not. Identifying and closing variance is the highest-return activity a cloud kitchen operator can perform.
How to Calculate the Variance
Theoretical food cost is calculated from your order records: for each order dispatched, sum the recipe costs of all items in that order. Multiply by order count for the period. This is what you should have spent if every dish was produced exactly to recipe, with zero waste, spoilage, theft, or error. It is not adjusted for reality — it is the mathematical ideal.
Actual food cost comes from your purchase records and inventory counts: beginning inventory value plus purchases during the period minus closing inventory value equals cost of goods used. If you do not do a physical inventory count at the end of each week, you cannot calculate actual food cost with accuracy. Estimating closing inventory produces an estimated variance — which is not useful for diagnosing specific problems.
Actual Food Cost = Opening Inventory + Purchases - Closing Inventory
Variance = Actual Food Cost - Theoretical Food Cost
Variance % = (Variance / Net Revenue) x 100
What the Variance Tells You
| Variance | Likely Causes | Investigation Steps |
|---|---|---|
| 0-2% | Normal kitchen variation | No action needed. Continue monitoring weekly. |
| 2-4% | Portion size drift, minor waste, minor over-orderingusage. | Reweigh two or three top-volume dishes. Check closing inventory accuracy. Review purchase orders vs actual |
| 4-6% | Consistent over-portioning, spoilage, recipe deviation | Mandatory spot checks on portioning. Identify items with highest absolute waste by weight. Check refrigerator temperatures and FIFO discipline. |
| Above 6% | Likely theft, systematic recipe deviation, or inventory counting error | Full audit of inventory count methodology. Install portion-scale spot checks for every shift. Review access controls. Recost the 5 highest-volume dishes from scratch using current supplier prices. |
At Rs 4,00,000 monthly net revenue, a 5% variance = Rs 20,000/month of unaccounted food cost. Over 12 months, that is Rs 2.4 lakh — enough to fund a second kitchen bay, a full equipment upgrade, or two months of marketing budget.
The single most common cause of high variance in Indian cloud kitchens is inconsistent portion weights during dinner rush hours when throughput pressure is high. A 15% over-portion on a 180g protein portion (27g extra) at Rs 280/kg costs Rs 7.56 per dish. At 35 orders/day, that is Rs 264/day, Rs 96,360/year — entirely from one slightly-too-generous ladle.
Cost Benchmarks by Cuisine Category
How food cost targets vary depending on what you cook
Food cost targets are not uniform across all cuisines. The ingredients, preparation complexity, and customer price sensitivity of different cuisine categories produce systematically different food cost structures. Understanding where your cuisine category sits on the food cost spectrum allows you to set a more accurate target and to evaluate your performance against the right benchmark.
| Cuisine Category | Typical FC % | Avg AOV (Rs) | Eff. FC % target | Key Driver |
|---|---|---|---|---|
| Biryani / Rice dishes | 28-33% | 320-480 | 35-42% | Protein cost (chicken, mutton) dominates. Portion weight critical. |
| North Indian (dal makhani, paneer) | 26-31% | 280-450 | 33-39% | Dairy (paneer, cream) drives FC. Veg dishes have better margins. |
| South Indian tiffin | 22-28% | 180-320 | 29-36% | Low ingredient cost, but lower AOV compresses absolute margin. |
| Chinese / Indo-Chinese | 25-30% | 250-400 | 32-38% | High-volume sauces and bases allow efficient production. Packaging heavier. |
| Pizza / Burgers / QSR | 23-28% | 250-450 | 29-35% | Standardised recipes with low waste. Strong upsell potential with sides/drinks. |
| Healthy bowls / salads | 28-36% | 350-600 | 35-44% | Expensive fresh ingredients, short shelf life. Higher AOV required to compensate. |
| Desserts / Mithai | 25-35% | 200-800 | 32-42% | Dairy and sugar dominate. Premium presentation allows higher pricing and better margin. |
QSR and fast food categories (pizza, burgers) consistently achieve the lowest food cost percentages because standardised recipes enable precision portioning, shared ingredient bases minimise variety, and high order volumes allow efficient batch production.
Healthy bowls and salad categories have the highest food cost risk because fresh ingredients carry short shelf lives and high spoilage potential. Operators in this category need the tightest inventory management and the most frequent stock rotation.
South Indian tiffin has the lowest absolute food cost but requires high order volume to generate meaningful absolute profit because the AOV is lower. A tiffin kitchen needs 50+ orders/day to generate the same absolute profit as a biryani kitchen doing 30 orders/day at higher AOV.
Six Food Cost Mistakes to Fix This Week
Common errors that inflate food cost percentages — and how to correct them
Most cloud kitchens that are underperforming their food cost targets are not making a single dramatic error — they are making several small errors simultaneously, each contributing 1-3 percentage points of inflated food cost. The cumulative effect is a food cost that is 6-10 points above where it should be. The six errors below cover the most common culprits.
Mistake 1 — Costing Without Yield Factors
Buying 1 kg of raw chicken at Rs 220/kg and costing dishes at Rs 220/kg is a systematic error. After butchering, skin removal, and bone, usable meat yield is typically 65-75% for bone-in cuts. The effective ingredient cost per kg of usable protein is Rs 293-338/kg, not Rs 220/kg. This error alone inflates recipe cost calculations for protein-heavy cuisines by 25-35%.
Mistake 2 — Ignoring Packaging in Food Cost Calculations
Packaging is a direct cost of delivery and must be tracked alongside food cost. A biryani packed in a foil tray with paper bag, heat seal, and sticker label costs Rs 18-25 in packaging per order — 4-7% of a Rs 380 average order value. Operators who track food cost but not packaging are missing a cost line that is as large as their monthly rent.
Mistake 3 — Using Platform GOV as Revenue in Food Cost Calculations
Food cost percentage calculated against gross order value — the amount the customer paid — dramatically understates the true food cost burden. Rs 100 of food cost on a Rs 340 order is 29.4% of GOV. After 25% platform commission, the same Rs 100 is 39.2% of net revenue. Food cost must be calculated against net revenue (what you actually receive) to be meaningful for margin management.
Mistake 4 — Weekly Stock Counts That Are Not Actual Counts
A stock count where the kitchen team reports estimated quantities rather than physical weights and counts is not a stock count — it is a guess, and it produces an unreliable actual food cost figure. Physical inventory counting takes 30-45 minutes in a well-organised cloud kitchen. This 30-minute weekly discipline is what makes variance analysis possible. Without it, food cost is a theoretical estimate, not a managed metric.
Mistake 5 — Not Updating Ingredient Prices Monthly
A recipe costed in January using Rs 35/kg tomatoes will produce an incorrect food cost calculation in June when tomatoes are at Rs 95/kg. Indian ingredient prices for fresh produce, dairy, and edible oils are highly seasonal. Operators who update their unit price master sheet only quarterly or annually are running on stale data for 9-12 months of every year. Monthly updates require 20-30 minutes and prevent silent margin erosion.
Mistake 6 — Not Separating Complimentary and Promotional Item Costs
If you send a complimentary sweet with every order as a brand gesture, or if Swiggy or Zomato runs a platform-side promotion that discounts your items, these costs do not disappear from your food cost — they inflate it. Track complimentary items separately as a marketing cost line rather than absorbing them into food cost. Platform-side discounts that reduce the effective selling price should trigger an immediate effective food cost recalculation.
Recipe cards exist for every menu item with gram-level quantities and yield factors.
Unit price master sheet updated from purchase invoices in the last 30 days.
Weekly physical inventory count completed on the same day and time each week.
Packaging cost tracked as a separate line from food cost.
Food cost % and effective food cost % calculated separately in your P&L.
Actual vs theoretical variance calculated and reviewed weekly.
Complimentary items tracked as marketing cost, not food cost.
Menu prices cross-checked against effective food cost (post-commission) annually.
Managing food cost with F&Bos
F&Bos costs every recipe down to the gram with yield percentages, deducts ingredients on every bill, and compares ideal with actual food cost so leaks show up the same week. MarginMind AI shows the margin on each dish at current ingredient prices.
Direct Zomato and Swiggy integration is on the roadmap, so record delivery orders in your cloud kitchen POS and work out effective commission with the Zomato & Swiggy commission calculator.