I suppose agreeing on terminology and collecting past historical data are crucial steps in developing a science. Historically collecting good data in economics

I suppose agreeing on terminology and collecting past historical data are crucial steps in developing a science. Historically collecting good data in economics has been painstakingly slow, especially in large scales. Maybe the field will move forward soon towards better explanations. I have no problem with the fact that human understanding is incomplete. What bugs me is that economic models of questionable utility are taken as gospel in politics and used as a battering ram - "because science told us to".