What owner-operators on nouz taught us about food cost ratios.
The median café runs 31,4% food cost. Top quartile sits at 27,1%. Here's where the difference comes from — and the four levers that actually move the number, based on aggregated data from cafés on nouz across twelve European countries.
Across European cafés running on nouz between January 2025 and March 2026, the median food cost ratio sits at 31,4% of net revenue. The top quartile clears 27,1%. The bottom quartile drags at 37,8%. That gap — almost 11 percentage points between best and worst — is the difference between a café that scales and one that grinds. Here's where it comes from.
Methodology
Anonymised, aggregated daily P&L data from cafés on nouz with at least 90 trading days in the sample window (Jan 2025 – Mar 2026). Twelve countries, predominantly Germany, Austria, Italy, France, the Netherlands, Portugal, Czechia. Food cost = COGS on food + beverage items, divided by net revenue (after VAT and card fees). Outliers (>50% or <15%) excluded as likely data-entry errors.
The spread: 27% to 38%
| Quartile | Food cost % | Implied EBIT (typical café) |
|---|---|---|
| Top (Q1) | 27,1% | €38.000 / year |
| Median | 31,4% | €24.000 / year |
| Bottom (Q4) | 37,8% | €6.000 / year |
The numbers above assume a €380.000-revenue café with stable rent and staff costs. The food cost line is the single largest swing factor in annual EBIT — bigger than any other variable cost. Move from median to top quartile and you find €14.000 a year. Slip from median to bottom and you lose €18.000. For the definition behind this ratio, see the food-cost percentage glossary entry; for the broader sector picture, see small-business profitability statistics 2026.
What separates the quartiles isn't the menu. It isn't the supplier. It's four operational habits.
Lever 1 — Supplier consolidation
Top-quartile cafés use a median of 4,2 suppliers. Bottom-quartile use 7,8. More suppliers means more invoices, smaller per-order volumes, weaker pricing power. The single biggest move a median café can make: kill the smallest three suppliers and consolidate volumes into the largest one.
Lever 2 — Portion drift
Top-quartile cafés show ±2% portion variance on key ingredients over a quarter. Bottom-quartile drift to ±9%. The drift is invisible day-to-day — it shows up as a 4-point margin slide over six months. The fix is mechanical: weigh portions on a kitchen scale once a week and compare to recipe spec.
Cafés that set up products with exact COGS in nouz get a daily portion-drift signal automatically — the margin on each cappuccino sale moves in real time, so an under-portioned latte from a tired Saturday is visible on Monday morning.
Lever 3 — Recipe seasonality
Top-quartile cafés re-cost their top five recipes quarterly. Bottom-quartile re-cost annually or never. Why it matters: butter, milk and seasonal produce move ±15-25% across a year in European wholesale. A recipe priced from January costs doesn't survive July.
See our detailed pricing walkthrough for the per-item exercise. For the why, see how daily prime cost surfaces drift before it hardens.
Lever 4 — Daily margin watch
Top-quartile cafés check their margin ≥4 days a week. Bottom-quartile check monthly or less. This is the meta-lever — it's why the first three habits stick. Without a daily check, supplier hikes go unnoticed, portion drift compounds, and recipe seasonality slides past. A weekly check catches most of it; daily catches all of it. The mechanics of the daily check sit in the daily P&L pillar; the café-specific version is the day-with-nouz walkthrough.
The 60-second daily routine we wrote up in a separate post is the operationalisation of this lever. It's why the routine matters — not for the routine's sake but because it makes the other three levers possible.
What to do this week
- Pull your last 90 days of food costs. Compute your food cost ratio. Where do you sit vs. the median?
- If above 32%: pick one of the four levers and act on it within seven days.
- If between 28% and 32%: you're mid-pack. The fastest move is portion drift (lever 2).
- If below 28%: you're in the top quartile. Keep doing what you're doing, and watch for drift quarterly.
If you don't have 90 days of clean daily P&L data, get started with nouz. By the time your first quarter is done, you'll know exactly where you sit on this curve. Food cost is half the story — the other half is labour, walked through in the café labour-cost benchmark, and the two roll up into daily prime cost as the operational headline number. For the full cross-vertical view of how this fits into a daily P&L, the cafe operator's master guide is the synthesis.
FAQ
What's a good food cost ratio for a small café?
European median is 31,4% of net revenue. Top quartile sits at 27,1%. Anything under 30% is considered healthy; anything over 35% is a red flag worth a closer look.
How does food cost ratio differ between specialty coffee and full-service cafés?
Specialty coffee shops tend to run lower (28-30%) because espresso has high margin and lower portion variance. Full-service cafés serving food run higher (32-36%) because plated food drifts more and costs more per gram.
Does this benchmark apply to bakeries?
Partially. Bakeries have higher ingredient COGS (35-42% typical) but lower labour-as-percentage. The four levers still apply but the absolute ratio you're aiming for is different.
How does nouz calculate food cost ratio?
Total COGS (snapshotted at the moment of each product sale) divided by net revenue (after VAT and card fees) for the period. The ratio updates daily and you can see the trend on the Statistics tab.
Where does food cost sit inside prime cost?
Prime cost = food cost + labour cost as a share of net revenue. Food cost alone is half the picture; the prime-cost glossary entry defines the combined number and restaurant prime-cost mastery walks through how to manage both lines together.
Where can I see more benchmarks for cafés like mine?
The small-business statistics hub aggregates the cross-shop dataset by sector and topic; the café profitability pillar reads them in plain language against an owner-operator P&L.