One thing I've noticed is someone who is very good at a tech job isn't just twice as productive as someone who is lousy at it; the discrepancy could easily be 10x; or it could be that he produces positive progress and the lousy guy produces anti-progress. This is clearly true for software developers, but I've seen it happen with network administrators too: small cadres of happy, super-productive admins outperforming armies of miserable tech drones.
But the thing is if you don't understand anything about (a) the technology or (b) human beings, how do you get a worker to be more productive? You make him work longer.
I'm not talking about striking while the iron is hot. When opportunity produces the occasional 80 hour work week, that's a totally different matter than having no better idea of what to do than setting unrealistic goals and leaving it to workers to make it up through sheer, unsustainable effort. Too often in the latter case you end up producing the semblance of progress. Yeah, I finished the module but someone's going to have to throw it out and rewrite when it blows up in the customer's face.