Start here
Cream cool butter, not soft butter
Cool room temperature is the target, around 67°F: it should give slightly under a finger and leave no grease on it.
Butter that is properly soft, let alone warm, has already started the spread before the tray reaches the oven. This is the most common invisible cause, because “softened butter” is written in every recipe and defined in almost none of them. Cooks who creamed visibly chunky, under-softened butter got a heavier spread too, so this is a narrow window rather than a one-way rule.
Then chill
Chill it, and expect it to do three things
Thirty minutes measurably reduces spread. It also makes the dough workable, which is the job cooks here actually reported needing.
Nine separate cooks hit dough too soft to roll, and kneading it back to pliable after a rest is what fixed it. The third effect is a trade rather than a gain: a chilled dough browns LESS. If you want colour on the edges, you are buying it back with time or temperature, not with the fridge.
Where chilling fails: when it is asked to rescue a formula. Cooks reported it working and not working at about two to one, and the failures cluster on doughs that were too soft or too rich underneath.
The measured lever
830 bakes
Sugar is what actually carries the dough outward
More total sugar tracked with a wider cookie across more bakes than anything else we hold on this dish.
Sugar melts before the structure sets and takes the dough with it, which is why a sugar cookie -- a dough that is mostly flour, sugar and fat -- spreads so readily. It is also why cutting sugar to control spread costs you the crisp edge and the pale gold colour at the same time. If you want the shape without losing the cookie, spend your effort on temperature and chilling first.
From 830 bakes. The best-evidenced thing on this page by a distance.
Do not do this
Do not add flour because the dough looks loose
A sugar cookie dough is supposed to look slack when you finish mixing it, and to firm up during the rest.
Adding flour on sight is the correction that bakes a dense, dry regret, and it is the single most common way people make this worse while trying to make it better. Extra flour does reduce spread: one clean test added a quarter-cup and the spread dropped to the bottom of the scale with nothing else changing. That is a real effect and it is not free, and it belongs in the recipe rather than in a panicked handful.
If you do need more flour: weigh it. One recipe in the corpus states a gram figure 14% above its own cup figure, and a scooped cup against a spooned one produced one of the widest cookies anyone reported.