Start here
Weigh your flour
A cup scooped straight from the bag can hold a third more flour than a cup spooned in and levelled, and that difference on its own is enough to turn a chewy cookie cakey.
This is the uncomfortable part, so we will say it kindly: the recipe is usually fine, and the measuring is not. Nobody gets it wrong on purpose. Scooping packs the flour down, the same cup line gives you a different cookie on different days, and flour is the ingredient a cookie is least forgiving about. A scale removes the whole variable. Set it to 120 g per cup and bake the recipe you already have before you change anything in it.
Honest note: recipes record what they asked for, never what went into your bowl, so this is the one cause we cannot put a number on. We are still putting it first, because it is still the most likely one.
Skip this one
18 bakes
Do not cut the baking soda
Cutting the leavening is the most repeated advice for a puffy cookie, and the cooks we read got the opposite result: more leavening made cookies spread flatter, not taller.
Leavening does two jobs at once. It lifts, and it shifts the dough toward alkaline, which weakens the structure that would have held that lift up. The rise you get early is paid back by a softer set later. Adjusting it moves your cookie in the direction you did not want, and not very far.
From 18 bakes where a cook changed the leavening and nothing else, and told us what came out.
Not the cause
Leave the eggs alone
Egg varies a lot across these recipes, and none of that variation showed up as a difference in spread.
Egg has a real job in a cookie: it firms the middle and browns the surface sooner. It is simply not the thing that decides how far the dough travels on the tray. Pulling an egg to fix puffiness costs you texture and does not buy you spread. Add a yolk if you want a richer cookie, but make that a flavour decision, not a repair.
Open question
27 of 619 use powder alone
If your recipe says soda, use soda
Several sites blame baking powder for cakey cookies, but almost no published recipe uses powder on its own, so there is nothing to compare it against.
Of the 619 recipes we resolved, 27 use baking powder on its own, which is not enough to compare against anything. The idea is reasonable, since powder brings its own acid and behaves differently in the dough. But reasonable is not the same as tested, and we would rather tell you that than dress a guess up as a finding.