Why? What are the worst consequences of wrong metrics? The problem there is that its a paradox. Seriously, consider what you are saying. You are saying measurements should not be conducted, because if somebody believes they are bad measures some mysterious unknown consequence will come to pass. The fallacy there is that the same consequences will come to pass irrespective of any measures conducted, because measurements alone do not dictate courses of action. However, if you don't conduct measurements you won't have any idea either way, which makes the very concern of bad measures entirely moot.

Bad measurements are only bad when they disqualify better measurements and better measurements only occur when some measure more directly proves/disproves the desired result.