When you scale a recipe up by 1.5× or more, the tool shows a few notes about leavening, and about bake time. Here is the reasoning behind them, checked against professional baking references rather than general kitchen folklore, and one thing we got wrong the first time and then fixed.
An earlier version of this tool told everyone to cut leavening to about three-quarters when scaling up. That was wrong, and here is why: professional baking uses baker's percentage, where every ingredient, leavening included, is a fixed share of the flour weight at any batch size. Under that standard, leavening scales in a straight line, the same as flour and sugar. There was no authoritative source behind the three-quarter rule; it turned out to be a figure repeated across recipe blogs without anyone tracing it back to an actual baking reference.
What is actually true is more specific than one blanket rule:
Baking soda is the one that genuinely needs care, but not because of the amount itself. Soda needs to react against an acidic ingredient (buttermilk, brown sugar, yogurt, cocoa). As long as that acid scales up in step with the soda, the soda is fine scaled the same way as everything else. Leftover, unreacted soda is what causes a soapy or metallic taste, not a large batch on its own.
Baking powder is self-contained (it carries its own acid), so it scales more predictably. If a scaled-up bake rises fast and then sinks, the original recipe was likely a little generous with powder to begin with, not too large a batch.
Yeast scales in proportion for any home-sized batch. The only real exception is genuinely large batches, well beyond a double, where a bigger mass of dough retains its own heat and ferments a little faster.
This one held up under checking, with one addition worth being precise about. Heat moves into batter or dough by conduction, and the time that takes follows the physics of heat diffusion: the time to heat through scales with the square of the thickness, not the amount. Doubling the depth of batter can roughly quadruple the time it takes to heat the center; doubling the quantity while keeping the depth the same barely changes it at all.
That is why the tool's guidance leads with depth, not volume: keep the same depth by using a bigger pan or a second pan, and bake time is close to unchanged, so start checking around the original time. Pour the same doubled batter into one deeper pan instead, and it typically needs roughly 15 to 25% longer, with the oven temperature dropped about 25°F so the outside does not finish well before the middle does. Either way, a toothpick, a thermometer, or how the bake looks is a better guide than the clock.
A calculator that just multiplies numbers back at you is not that useful. Advice that sounds right but was never checked against an actual baking reference is worse than no advice at all. This page exists so the reasoning behind what the tool tells you is checkable, not just asserted, and so a mistake we made and caught gets corrected in the open rather than quietly edited away.