Nice piece in the NY Times on Google's HR metrics and the effort to build a better boss. I have a section in IFL on one attempt to restructure management and the fallout from that experiment. There was definitely a bias in favor of tech knowledge over touchy-feely management sensibility, which worked better for some people than others.
The company was always trying to reduce the human resources puzzle to a solvable equation, and introduced metrics that were supposed to evaluate our skill as talent scouts by tracking the performance of candidates who were hired based on employee referrals. Presumably, if someone you recommended received good reviews once on the job, your opinion should be given more weight in the hiring process. Since no one I recommended ever got hired (not an uncommon occurrence), I suppose I could claim to have a perfect record on that front.