The annual get away from the cold trip was a great time with a ton of miles (done without too much fitness, at all), a bunch of statistics homework, a lot of recuperation in the hot tub, massive amounts of Mexican food, and more or less no time for writing blogs. So we'll bring this one to the finish line here.
The big thing at this point is to try and stress the wheel at least as much as the user will do (barring catastrophe) so that any bad behaviors of the build will identify themselves and get corrected. There's a lot of variation in how much of this you have to do, and I've never gotten a super tight bead on why. Some rims, I know there's going to be some fighting. Others, there almost never is. But you get surprises in each direction. You just keep stomping on the thing until further rounds of stress fall to either reduce spoke tension or take the wheel out of true.
(I'm refinishing my stairs)
And now we kind of go back to the beginning, because the big variable once it leaves the stand and goes into a box is "is the wheel well designed and spec'd for the use parameters?" If we send out a 20/24 build with light hubs and spokes, and under-gunned rims to a rider with exceptional power and torque, or for a use that's going to include a lot of severe pucker factor riding, we've messed up. The best built wheel for that torture chamber will last a lot longer than what could happen with a lesser build, but it's on borrowed time from the start in any case. On the other hand, good selections along with a good build shouldn't need much if any adjustment over the life of the wheel (which itself varies a ton), except in exceptional circumstances.
So let's invert things, and ask why wheels go out of true. Following on what's been said before, inadequate stress relief, unbalanced spoke tensions, round errors, and inadequate spec are usual suspects. And yet still well built, well spec'd wheels go out of true. Why?
Damage is a big one. The number of wheels we see with damaged heads-in drive side spokes is... high. What's happened there is that the chain was allowed to go over the cassette and get jammed between the chain and cassette. This beats the crap out of the spokes, bending and nicking them. Alloy rims that spend a lot of time in the air and less time landing from said time in the air can suffer flat spots. You can't take a wheel out of round without also taking it out of true, so that happens. Generally carbon wheels resist this, as carbon don't bend.
Really high loads, such as happen when you crash and someone rides over your wheels, or when your wheel gets stuck in a rut or thaw crack can cause an "over threshold" event and elongate a spoke or put a bit of a stretch in the rim.
With lesser hubs, the spoke holes can elongate over time and this takes the spokes out of tune. This is a thing we try HARD to take to its maximum during the build, we think about it as doing a "bench forging," if you will, of the spoke holes. Really lesser hubs, the whole flange can oblong. You'll also be ripping through bearings if that happens.
So a very long story about what makes wheels true, and helps them stay that way, and causes them to not stay that way. These are all things we have in mind when we're building wheels. Though things don't always go according to plan 100% of the time, we aim for 100%. Machines can't, so never say never but there's no machine built wheel that we've seen that can compete with a hand built one in heavy use. But not every hand built wheel goes through all this stuff that we and other very conscientious builders do, so there's a big range there, as well.