How to Scale a Cake Recipe

Every factor below is the target pan's batter volume divided by the original's, computed from the pans' actual dimensions rather than looked up from a chart written for different pans.

1. Find the batter volume of the pan the recipe was written for and the pan you actually want to bake in, from a batter volume chart or calculator, then divide the second by the first to get the raw scale factor.
2. Multiply every ingredient by that factor, in weight rather than volume wherever the original recipe allows, since scaling a cup measurement compounds inaccurately across a whole recipe.
3. Round the flour, sugar and fat quantities to the nearest gram or a workable fraction, since a kitchen scale reads to the gram but a factor like 1.56 rarely lands on one.
4. Crack the original recipe's number of eggs times the scale factor, whisk them together in one bowl, weigh the total, and pour out only the proportional weight rather than trying to divide a whole egg by eye.
5. Taper chemical leavening back slightly once the factor passes about two, holding baking powder or baking soda to roughly a tenth less than the straight multiplication would call for.
6. Check the batter against the new pan's two-thirds fill line before pouring, since a scaling error shows up here, as a pan that is obviously too full or too empty, well before it shows up in the oven.
7. Expect the bake time to move with the new batter depth, not just the new pan's width, and start checking at the low end of a fresh computed estimate for the new pan rather than reusing the original recipe's time.
8. Bake a small test batch first if the factor is unusual, such as scaling into an odd number of pans, since a test bake is far cheaper than a failed tier the night before an event.

Chemical leavening works by releasing gas into batter that has not yet set, and the crumb structure has to hold that gas in as small, even bubbles rather than letting it force its way out or collapse the structure before baking finishes. In a small batch that balance is forgiving; in a much larger batch scaled at the full straight multiplication, the same proportional gas release happens inside a bigger, slower-setting mass of batter, which gives the bubbles more time and more room to over-expand, coalesce, or blow out the surface before the structure is ready to hold them.

The usual practice is to hold baking powder or baking soda back by roughly a tenth once the scale factor passes about two, rather than multiplying it at the same rate as flour and sugar. This is a working adjustment based on how the trade handles large batches, not a fixed formula, and it is worth nudging on the next bake if a scaled cake still domes hard or has an uneven crumb despite the adjustment.

Because whole eggs are not a fine enough unit for most scale factors. A recipe that calls for two eggs, scaled by a factor of 1.27, genuinely needs about 2.5 eggs, and there is no accurate way to crack half an egg by eye. Whisking the needed whole eggs together in one bowl first, weighing the total, and then measuring out only the proportional weight the scale factor calls for solves this cleanly, and it works for any factor rather than only the ones that happen to land on a whole number.

A useful working figure is that a large egg without its shell weighs close to 50 grams, split roughly 30 grams of white to 20 grams of yolk, though real eggs vary enough that weighing the actual eggs you cracked beats assuming that figure. This matters most on a downscale, where counting a fraction of an egg by eye is genuinely difficult, and least on an upscale by a round number, where whole eggs still divide cleanly.