It seems like you said that disdainfully/sarcastically, but if you had metadata about thousands of experiments and the ability to do a meta-analysis on that data; wouldn't you? It seems like a great opportunity to identify potentially destructive trends in the way your company runs experiments.