I make only goat's milk soaps (for now), and my experience with gelling has been mixed, at best:
If it DOES go to full gel, then about 50% of the time, it comes out fine. About 30% of the time, it gets spongy/crystallized in the top 1/3 of the log, toward the center. The remaining 20% of the time, it separates, and I become very sad, like Eeyore, for a couple of days (especially if it had patchouli in it... that just hurts). Yesterday's batch falls into this last category, and today I am sad. I don't re-batch separated batches, because they are stinky.
It also seems that slight overheating is the condition that causes the separation, but the most powerful overheats are what cause the crystallization. If you're gonna blow it, go way over the top! It's ugly, but at least you can use the spongy stuff. But badly separated soaps are just nasty.
However, if it does NOT gel, or only partially gels, where there is about a centimeter of un-gelled skin on top, then it always comes out perfect. Always. Because of this, I leave it uncovered for the first couple of hours, but that's not always enough.
So, I don't like to see it go to full gel, and I get nervous about partial gel that almost gets through the skin. I think milk soaps are more prone to separation problems when they gel, but I have not yet found out why.
I'd really like some input from the experts here on separation, as related to heat, liquid percentage, milk usage, and degree of trace at pour. Yesterday's batch wasn't any hotter than normal (soaped at about 115F), but I did have a little more liquid than usual (25% total milk and water, I do split batch, and usually go with 22% liquid - not a big difference, but maybe enough), and poured at a thin-to-medium trace. I was trying to counteract cinnamon seize, but ended up with a total loss of 10 lbs of goodness. The world is a poorer place...