Why scaling is not just multiplication

The intuitive approach to scaling a recipe is to multiply every ingredient by the same factor. That approach works well for soups, stews, salad dressings, and most stovetop cooking. It fails, sometimes spectacularly, in baking.

The fundamental reason is that baking is chemistry, not arithmetic. Flour, fat, sugar, eggs, and liquid interact through a tightly balanced set of physical and chemical reactions. Each ingredient plays a structural role.

When you double a recipe, you are not simply making more of the same product. You are scaling an entire chemistry experiment, and some variables in that experiment do not respond to scale the way you expect.

The three categories of scaling behavior

Every ingredient in a recipe falls into one of three categories.

CategoryBehaviourWhat falls here
LinearScales directly with the multiplierFlour, sugar, butter, oil, eggs (mostly), dairy, water, extracts, chocolate chips, nuts, fruit
Sub-linearScaled to less than the full multiplierBaking powder, baking soda, salt, strong spices (cayenne, cloves, cinnamon)
Non-linearCannot be scaled by multiplication at allBaking time, pan size

Use the full multiplied amount on a sub-linear ingredient and you get over-leavened, bitter, or unpalatably salty results.

Non-linear is the awkward one. A doubled batch in the same pan does not need twice the baking time, it may need only 10 to 20% more, and pan size math uses area rather than volume.

The professional kitchen rule

Restaurant cooks and professional bakers treat seasoning (salt, acid, heat) as something added "to taste" at the end, never blindly scaled.

The same principle applies when scaling at home. Your calculated amounts are a starting point, so always finish by tasting.

Where cooking and baking differ

In cooking (soups, stews, sautées, braises), ingredients interact through straightforward flavor combination and heat transfer. Scaling a chicken broth by 3x makes 3x more broth.

The only watch-out is evaporation. A larger pot has more liquid but proportionally less surface area, so evaporation-based reduction happens more slowly.

If your original recipe relies on a long reduction, a tripled batch may need slightly less cook time, or you may want to reduce it separately.


In baking, structure is everything. Gluten networks, protein coagulation, sugar caramelization, and gas bubble formation all interact.

Over-scale leavening and your cake rises too fast, the gluten structure cannot support the expansion, and the center collapses into a gummy well. Under-scale and you get a dense brick.

Scale salt incorrectly and you taste it (too much) or notice the blandness (too little, since salt also suppresses bitterness in sweet baked goods).

How to scale any recipe, step by step

Calculate your scaling factor

Divide desired servings by original servings. Say a recipe serves 6 and you need 15 servings. That is 15 ÷ 6 = 2.5×, and it becomes your base factor for every linear ingredient.

Identify ingredient categories

Go through every ingredient and tag it. Linear (most things), sub-linear (leavening, salt, strong spices), or non-linear (time, pan size). This takes one minute and prevents the most common mistakes.

Scale the linear ingredients

Multiply each linear ingredient by your factor exactly. At 2.5×, 2 cups flour becomes 5 cups.

3 large eggs becomes 7.5 eggs. Use 7 eggs plus half a beaten egg, which is what the calculator on this page returns for the same input. A large egg out of the shell is close to 50 g, so half is about 25 g on a scale.

Convert fractional units to practical measures using a cups-to-grams converter where helpful.

Apply the sub-linear rules

Four groups come off the base factor instead of following it.

Ingredient groupStart at
Leaveners and thickeners (baking powder, baking soda, cornstarch, arrowroot, gelatin)75%
Flavorings (vanilla, almond, herbs)67%
Strong spices (cayenne, cloves, nutmeg)60 to 70%
Saltto taste, 70% as a start

The 75% and 67% figures are the Culinary Institute of America's.

Salt is the odd one out. Most sources add it to taste rather than scaling it, and My Curated Tastes puts the numeric starting point at 60 to 70% of the scaled amount.

Worked at 2.5×, that is 0.75 × 2.5 = 1.875×, so 1 teaspoon of baking powder becomes roughly 1⅞ teaspoons.

Adjust pan size and baking time

Choose a pan with an area that matches your scaling factor times the original pan area. Round pans use area = π × r², rectangular pans use area = L × W.

Bake at the same temperature, but start checking for doneness 10 to 15 minutes before the original recipe's end time. Use a toothpick or instant-read thermometer rather than relying on time alone.

Why weight beats volume when you scale

Scaling becomes far more accurate when you work by weight (grams) rather than volume. A cup of flour can vary by 20% depending on how it was scooped.

Use our Cups to Grams converter to build a weight-based ingredient list before you scale, and the math becomes trivial multiplication.

Five scaling mistakes, and the exact fix for each

Mistake 1 Scaling baking time linearly

Doubling a recipe and doubling the baking time is the single most common error. Baking time is determined by heat penetration depth, not food volume.

A thicker batter in the same pan bakes from the outside in. The center temperature lags the edges because heat diffusion slows with depth.


So baking time changes far less than batch size. A 2× batch in the same pan usually needs only 10 to 20% more time, and two pans of the original size need almost the same time as one pan.

Always test with a toothpick (inserted in the center, it should come out clean with no wet batter) or a thermometer (internal temp 200 to 209°F for cakes and cupcakes, 200 to 205°F for quick breads, per ThermoWorks).

Rule: same pan depth = same time (±10%) Deeper pan = add 10-20%, check early Two pans side by side = same time (ensure airflow)

Mistake 2 Scaling salt to the full multiplier

Salt perception is not linear. Humans detect salt concentration (parts per million relative to the total flavor mass), not absolute amount.

A doubled recipe does not need double the salt to taste equally salty, because the ratio of salt to other ingredients already reaches the palate at the same rate per bite.

Salt in baking also does two jobs that want less of it, not more. It suppresses bitterness in cocoa and slows yeast fermentation in bread.


So treat it the way the pros do. Cooklang treats salt as something you add to taste rather than scale mathematically.

If you need a numeric starting point, My Curated Tastes puts salt at 60 to 70% of the scaled amount, so about 0.70 is a sensible start.

Cooklang's own example lands in the same place. It takes 1 teaspoon at 2 servings to 2.5 or 3 teaspoons at 8, which is 0.63 to 0.75 of the linear amount.


For yeasted bread, where salt also slows fermentation, stay nearer the top of that range and watch the proof.

Starting salt ≈ calculated amount × 0.70 (or to taste) Then taste, and adjust up if it needs it

Mistake 3 Scaling leavening fully, and the CIA 75% rule

The most technically interesting scaling error happens with leaveners. Baking powder and baking soda create CO2 gas bubbles.

Use the full multiplied amount and the batter leavens too aggressively, the bubble structure stays unstable, and the center collapses during baking. The cake comes out sunken, with a gummy interior under a set exterior.


The published guidance is consistent here. The Culinary Institute of America puts thickeners and leaveners at about 75% of the calculated amount when scaling up. Cooklang reports the same 75% rule for baking powder and baking soda. Start there, then adjust upward if the rise is weak.

Thickeners like cornstarch, arrowroot and gelatin follow the same 75% rule.

Some baking writers, including Serious Eats contributor Stella Parks, scale leavening linearly at home-batch sizes and report good results, so the rule is most useful above 2× where over-leavening really shows up.


Multiply leavening and thickeners by 0.75 of your scaling factor as the starting point, evaluate the rise, and add more if needed.

Starting leavener = original amount × (scaling factor × 0.75) Example: 1 tsp baking powder, scaling 4× New amount = 1 tsp × (4 × 0.75) = 1 tsp × 3 = 3 tsp (Not 4 tsp) Example: 2 tsp baking powder, scaling 2.5× New amount = 2 tsp × (2.5 × 0.75) = 2 tsp × 1.875 ≈ 3¾ tsp
75% multipliers for common factors
Scaling factorLeavener multiplier
½××0.50
×1.50
×2.25
×3.00
×4.50
×6.00

Mistake 4 Choosing the wrong pan size

The batter depth in the pan determines baking time, crust-to-crumb ratio, and texture.

A pan with twice the area but the same batter depth produces the same baking time. A pan with the same area but twice the batter depth produces a significantly longer baking time and a denser crumb.


Most people think "bigger recipe = bigger pan" without doing the area math.

A doubled cake batter poured into one large pan may have much greater depth than the original, causing an undercooked center even after extended baking.

Match area to your scaling factor, and aim to keep batter depth within 15 to 20% of the original. See the pan size table below.

Round pan area = 3.14159 × (diameter / 2)² Square pan area = side × side Rect pan area = length × width Target new pan area = original area × scaling factor

Mistake 5 Ignoring liquid ratios in soups and stews

For stovetop recipes, scaling up creates a larger volume of liquid with proportionally less evaporation surface relative to volume. A wide shallow pan loses moisture fast, a narrow tall pot loses it slowly.

If your original recipe calls for reducing a broth to concentrate flavor, a tripled batch will take significantly longer to reduce to the same concentration, or it will never reduce fully.

The inverse problem occurs when scaling down. Less liquid in the same pan can reduce too quickly, burning or becoming too concentrated.


When scaling a soup or stew that relies on evaporation, use a wider pan to keep a similar surface-area-to-volume ratio. Or reduce the liquid separately in a skillet, then add it to the main pot.

Scaling down, watch closely and add small amounts of water or stock if it over-reduces.

Scaling up → use wider pot, or reduce liquid separately Scaling down → watch closely, add stock as needed Target: same surface-area-to-volume ratio as original

Scaling reference table

The table below shows exact multipliers for the 10 most common scaling scenarios, including the special rules for leavening agents.

Read the leavening column for baking powder and baking soda, and the base column for everything else.

Goal Base Factor (most ingredients) Leavener & Thickener (CIA 75% above 1×) Salt Starting Point (70% of scaled above 1×) Example: 1 cup flour Example: 1 tsp baking powder
Quarter (¼×) 0.25 0.25 0.25 ¼ cup ¼ tsp
Half (½×) 0.50 0.50 0.50 ½ cup ½ tsp
Two-thirds (⅔×) 0.67 0.67 0.67 10⅔ tbsp ⅔ tsp
Three-quarters (¾×) 0.75 0.75 0.75 ¾ cup ¾ tsp
Same (1×) 1.00 1.00 1.00 1 cup 1 tsp
Double (2×) 2.00 1.50 1.40 2 cups 1½ tsp
2.5× 2.50 1.88 1.75 2½ cups 1⅞ tsp
Triple (3×) 3.00 2.25 2.10 3 cups 2¼ tsp
Quadruple (4×) 4.00 3.00 2.80 4 cups 1 Tbsp
Six times (6×) 6.00 4.50 4.20 6 cups 1 Tbsp + 1½ tsp

When scaling UP, leaveners (baking powder, baking soda) and thickeners (cornstarch, arrowroot, gelatin) start at roughly 75 percent of the base factor, which is what the Culinary Institute of America publishes and adjust upward if the rise is weak. At or below 1× they stay proportional. Cooklang publishes the same figure, "Baking soda / powder, yeast ~75% of linear". The underlying reason is a ceiling on flour weight, and Bakerpedia puts a typical cake at 3 to 5 percent baking powder on flour weight, with too much giving a bitter taste and surface blistering.

Cooklang's published example scales UP, 1 tsp of salt for 2 servings becoming 2.5 to 3 tsp at 8 servings, which is 0.63 to 0.75 of the linear amount, so this page starts salt at 70 percent above 1×. Scaling DOWN is a different case and the tool does not pull salt back there. My Curated Tastes argues the opposite direction for a smaller batch. Writing about cooking for two, it advises starting salt at 60 to 70 percent of the scaled amount rather than at half, tasting as you go and adding more only if needed, because salt perception is not linear in volume and a smaller pan reduces faster. This page computes salt proportionally below 1× rather than encoding a curve no source publishes, and says plainly that a halved recipe usually wants a little more salt than the arithmetic suggests. Taste near the end either way. For context on how little salt is involved, Bakerpedia puts recommended usage at 1.8 to 2 percent baker's percent in bread and 0.75 to 1.25 percent in cookies and cake.

Always taste the finished dish and adjust upward as needed.

Pan size conversion

Pan area determines how much batter fits at the same depth.

When you scale a recipe, find a pan (or combination of pans) whose total area equals or closely approximates your target area (original area × scaling factor).

Area formulas

PanArea formulaWorked example
Roundπ × r², r = diameter ÷ 28-inch, 50.3 sq in
Squareside²9-inch, 81 sq in
RectangularL × W9×13, 117 sq in
LoafL × W9×5, 45 sq in
Bundt, 10 cupmeasured at the base58 to 65 sq in
Pan Dimensions Area (sq in) Cups Batter Common Equivalent Swaps
Round 8-inch 50.3 4 to 6 Two 6-inch rounds; one 9×5 loaf
Round 9-inch 63.6 6 to 8 8-inch square; two 6-inch rounds + extra
Square 8×8 inch 64 6 to 8 9-inch round; 9×5 loaf ×1.4
Square 9×9 inch 81 8 to 10 Two 8-inch rounds (slightly over)
Rectangular 9×13 inch 117 10 to 14 Two 9-inch rounds; two 8-inch squares
Rectangular 11×15 inch (jelly roll) 165 14 to 18 Three 9-inch rounds; two 9×13 pans (nearly)
Loaf 9×5 inch 45 4 to 6 8-inch round (slightly over); three 5.75×3 mini loaves
Tube / Bundt 10-inch ~65 base 10 to 12 Two 9-inch rounds; one 9×13 (roughly)
Pan substitution, worked

You have a recipe for one 9-inch round layer cake and want to make a sheet cake for 30 people, roughly 3× the recipe. The original area is 63.6 sq in, so the target is 63.6 × 3 = 190.8 sq in.

A 12 by 16 inch sheet pan gives 192 sq in and lands almost exactly there. One pan, same batter depth, same baking time range.


Watch the naming here. A true half-sheet pan is 13 by 18 inches, which is 234 sq in, about 22% more area than this target.

Pour the same batter into one and it sits shallower and bakes faster, so check it early rather than trusting the original time.

Recipe scaling questions, answered

No. Most ingredients do, like flour, butter, eggs, liquid, sugar, extracts, and mix-ins. Three categories do not.

When you scale UP, leaveners and thickeners (baking powder, baking soda, cornstarch, gelatin) start at about 75% of the calculated amount, per the CIA. At or below 1× they scale straight down, because that rule exists to stop a bigger batch over leavening and the reasoning reverses when the batch shrinks.

Scaling up, flavorings (vanilla, almond, herbs) sit at about 67% and strong spices (cayenne, cloves, nutmeg) at 60 to 70%. Scaling down, both stay proportional.


Salt goes to taste, not to math. Baking time and pan size depend on batter depth and pan area, not on the amount of food. The calculator above applies the sub-linear rules automatically when you pick "Baking" as the recipe type.

Multiply most ingredients by 2, then back off three things.

For baking powder, baking soda, cornstarch, and gelatin, start at 1.5× the original amount (75% of the 2× factor) and adjust upward if the rise is weak.

For flavorings like vanilla and almond extract, go to about 1.33×. Salt goes to taste rather than doubled.


Pick a pan with roughly double the area of the original. Two 9-inch rounds instead of one, or one 9×13 instead of a 9-inch round. Check doneness with a toothpick or thermometer, not by scaled time.

Use a scaling factor of 0.5. Most ingredients halve directly.

Baking powder, baking soda and thickeners halve directly too, at 0.5×. The Culinary Institute of America 75 percent pull back is a scaling UP rule, because leavening becomes over effective as batter volume grows. Nothing about that reasoning holds when the batch gets smaller, so applying it downward would leave a half batch under leavened.

For a single egg that cannot be halved cleanly, beat one egg and weigh out half of it, about 25 g. A large egg out of its shell is close to 50 g.


Baking time depends on batter depth. Halve the batter into a shallower pan and time can drop 15 to 20%.

Keep the same depth in a smaller pan and time is roughly unchanged. Always check early with a toothpick or thermometer.

Leavening agents create CO2 gas bubbles. When you scale a recipe up, the batter mass grows faster than the leavening's effective lift.

Use the full multiplied amount and you get over-leavening. The cake rises too fast, the gluten network cannot support the rapid expansion, and the center collapses, leaving a dense, gummy interior under a cracked top.


The Culinary Institute of America puts thickeners and leaveners at about 75% of the calculated amount when scaling up, and Cooklang reports the same figure for baking powder and baking soda.

Some baking writers, including Serious Eats contributor Stella Parks, scale leavening linearly at home-batch sizes below 2× without issue.

Over-leavening is the most common cause of sunken cakes in scaled recipes above 2×.

Calculate the area of each pan, then divide.

Round pan. Area equals π × r², where the radius is half the diameter. Square or rectangular pan. Area equals length × width.


Divide the new pan area by the original pan area to get the scaling factor. An 8-inch round (50.3 sq in) to a 9×13 pan (117 sq in) gives 117 ÷ 50.3 = 2.33×.

Plug that factor into the calculator above to get every adjusted ingredient amount.

No, and assuming they do is the most dangerous shortcut in recipe scaling.

Baking time is governed by heat transfer through the food mass, not the amount of food.

A doubled recipe in the same pan may need only 10 to 20% more time. A doubled recipe split into two same-size pans may need nearly identical time to the original.


Always use a thermometer (cakes and cupcakes are done at 200 to 209°F internal, quick breads at 200 to 205°F, per ThermoWorks) or the toothpick test, never scaled time.

Scaling UP, start strong spices such as cayenne, ground cloves, nutmeg and cinnamon at 60 to 70 percent of the calculated amount, and softer flavourings such as vanilla, almond and most herbs at about 67 percent. Then taste and adjust. Scaling down, both stay proportional. Those are the same figures the calculator on this page uses. Flavor compounds concentrate differently than bulk ingredients.

It matters most with strong spices like cayenne, cloves, cinnamon, and nutmeg, which can overwhelm a dish if over-scaled.

Fresh herbs are more forgiving than dried. The tool groups herbs with the flavourings at about 67 percent of the factor when scaling up, and dried spices and ground chiles sit lower, at 60 to 70 percent. Taste before you add more.

Yes, but tripling takes more planning than doubling.

For baking, the CIA rule puts leaveners at 2.25× the original (75% of the 3× factor) as a starting point.

Flavorings go to about 2.0×, which is two-thirds of 3×. At 3×, salt starts at 2.10× and strong spices at 1.95×, which is 0.70 and 0.65 of the factor, and both then go to taste. Those are the figures the reference table and the calculator both use.


You will almost certainly need multiple pans. Plan for three of the original pan size, or find a pan with triple the area.

For stovetop recipes, tripling can extend reduction time significantly due to the surface-area-to-volume issue. Consider using a wider cooking vessel or reducing liquid components separately.


Our recipe scaler on the homepage handles ingredient-by-ingredient scaling for large batch recipes.