How Do You Scale a Recipe — Without Ruining It?
Recipe scaling explained by a chef: the conversion factor, how to scale from the ingredient you actually have, how to hit a target batch weight — and what refuses to scale linearly.
To scale a recipe, multiply every ingredient by a conversion factor. The factor can come from servings (desired ÷ original), from the ingredient you actually have on hand, or from the total batch weight you need to produce. The arithmetic is simple. Doing it under pressure, on paper, for a twelve-line recipe, twice a day — that's the part every chef knows too well.
The daily reality: nobody scales for fun
It's 14:30. Your sorbet base recipe is written for 10 kg of raspberry purée. You open the walk-in: 8.5 kg — garde manger took 1.5 kg for a dressing this morning. Now the whole recipe has to move.
Pen and paper: 8.5 ÷ 10 = 0.85. Now multiply every single line by 0.85 — the sugar, the glucose, the stabilizer, the lemon juice — without slipping a decimal, in a kitchen that isn't waiting for you. Twelve ingredients, twelve chances to be wrong, and the stabilizer is the one line where "roughly" makes a grainy sorbet.
That's the real scaling problem. It's not that the math is hard. It's that it's constant, error-prone, and always urgent — and in a world where every chef carries a computer in their pocket, doing it on the back of an order sheet is a choice, not a necessity.
The three ways a factor is born
1. From servings — the textbook case:
Factor = desired servings ÷ original servings
Recipe for 4, need 10 → factor 2.5. Fine for à la carte. Almost never how production actually works.
2. From the ingredient you have — the walk-in case:
Factor = available amount of key ingredient ÷ recipe amount
The 8.5 kg purée: 8.5 ÷ 10 = 0.85 → every line × 0.85. This is how kitchens really scale: anchored to the main or second-main ingredient — the purée, the protein, the chocolate — because that's the ingredient you can't stretch and won't waste. The rest of the recipe follows it.
3. From the output you must hit — the "KC wants 6 liters" case:
The sous chef says: make 6 liters of dressing. Your recipe makes… how much, actually? Here's the trap: you can't scale to a target if you don't know your recipe's total yield. A recipe is a list of inputs; the factor needs the output:
Factor = required output ÷ total batch yield of the base recipe
If your base dressing recipe totals 2.4 kg (and roughly 2.4 L for a dressing near water density), then 6 ÷ 2.4 = 2.5 → every line × 2.5. No total on the recipe card? Then before any scaling happens, you're summing twelve lines by hand just to find your denominator. Write the total batch weight on every master recipe — it's the single most useful number on the card.
Try it in DigitalKnife →Open the web tools · free trial for pro plansWhat does not scale linearly
The factor moves every line — but the food pushes back. This is what separates a scaled recipe from a ruined one:
| Component | Scaling behavior | Rule of thumb |
|---|---|---|
| Salt & spices | Intensify faster than volume | Start at ~75% of the scaled amount, season to taste at the end |
| Chili, garlic, strong aromatics | Same — they compound | 70–80% of scaled, adjust late |
| Leavening (yeast, baking powder) | Large batches ferment more efficiently | ~80–90% of scaled for big multiples; watch the first batch |
| Gelatine & hydrocolloids | Set by concentration, not portion count | Scale exactly by liquid weight — precision items, no eyeballing |
| Wine, vinegar, reductions | Pot geometry changes reduction | Scale the liquid, judge the reduction by taste and consistency |
| Finishing herbs | Perception saturates | ~80% of scaled, finish by eye |
The pattern: anything whose job is flavor intensity scales sub-linearly, because taste isn't linear. Anything whose job is structure — gelling, binding, rising — scales exactly with the mass it acts on, and punishes creativity. In the sorbet, the purée scales at 0.85 and so does the stabilizer — to the gram.
Times, pots, and cooling don't scale at all
A 2.5× batch is not 2.5× the cooking time — heat transfer follows surface area and depth, not the factor. Deeper pots cook slower and reduce less; loaded ovens bake slower; and a 20-liter batch holds dangerous heat for hours after a 2-liter batch is already cold. When you scale significantly, treat the first run as a trial and write down the real times — that note beats any formula.
Scaling down bites differently
Halving has opposite traps: quantities fall below what equipment can handle, one egg becomes "half an egg" (weigh it — ~50 g large, so 25 g whisked), small batches reduce and scorch faster. Below roughly a third of the original, re-test — don't just re-multiply.
Related questions
Should I scale by servings or by weight? Professional kitchens scale by output weight ("6 kg of dough", "6 liters of dressing"), not servings — it's unambiguous. Which means every master recipe needs its total batch yield written on it.
What's the "key ingredient" method? Anchor the factor to the main or second-main ingredient — the one you have a fixed amount of (the purée, the protein) — and scale everything else to follow it. It's how kitchens avoid waste when stock on hand doesn't match the recipe.
Can every recipe be scaled? Almost — but emulsions, laminated doughs, and delicate pastry depend on technique-per-batch as much as ratios. Scale those by repeating original-size batches instead of one giant one.
Why did my scaled bread over-proof? Big doughs generate and hold more heat, so fermentation accelerates. Pull the yeast toward 80–90% of scaled and watch temperature, not the clock.
Catalin Cucu is a chef and the founder of DigitalKnife — where the Recipe Book scales any recipe live: change one ingredient's amount or the total yield, and every line recalculates instantly. The 8.5 kg problem, solved in two taps.