taras0000 Very helpful, thanks.
Cycles are tough to do without specific machinery. For the fully custom parts (as in, starting from a mold and building up from scratch) I was thinking I might send one to the ACT lab for stem/bar fatigue tests.
For some of the "semi custom" parts (that is, taking an existing whatever and modifying it, like narrowing drop bars or combining stem/bars) I don't think that would help as validation, since even if you made 2 of the same thing, there would be innate differences that you couldn't control/account for (curing, initial damages due to cutting/bonding) from 1 to 2. I suppose that's more akin to a repair than in that regard. Not sure if that's possible to really test repeatably. (and if you go too far testing a 1 off... it becomes unusable from the stress of the testing)