Sometimes the useful thing is not the best email, it is thirty different angles laid out next to each other so you can see which direction is actually worth committing to.
Deciding a direction. When a campaign angle is genuinely open, arguing about it in the abstract is slower and worse than reading thirty concrete versions and noticing which three make you sit up.
It is explicitly not the skill you build a campaign with. There is no ranking, no scoring gate and no sequence logic, and that is the trade that makes it fast.
A ranked list gets read down to about number four. The whole value here is the spread, including the angles that look wrong at first and turn out to be the interesting ones. Ranking would quietly delete that.
When a variation is clearly right, it can be promoted into a self-contained draft campaign with a full sequence and a campaign folder behind it. At that point it enters the normal process, with the gates and the scoring, rather than skipping them because it started here.
Skills compound. These are the ones we usually install alongside it.