Hans, I didnt get to read everyone's responses, but since this was posted 2 months ago, has anything changed since then?
It's a lot easier to just literally hold a contest of who draws the most accurate and fastest (compensation of some sort helps). Here in the east coast of all the firms I've worked for in the past, you have to perform to keep your job and know how good and fast your drafters are (and that's how you know who are the ones that help the company, and others are frankly "deadweight" or worse, "profit killers"). Any good drafter will see and know just by looking at what the final product should look like can tell you a rough estimate (even if it is off by 20%).
Productivity can't be measured of how much you do (as some can draw the same product and do it in less steps, but end up taking a longer amount of time, so really who is the most productive?). You have to have implement some kind of benchmark internally to know what your drafters can do (regardless of how complex the drawings can get, there's still fundamentals such as how it's still lines on a paper when said and done).
You really need to set aside at least 1 hour to have everyone draw the same object without them knowing that it is a part of a test. If they all end up using up the entire hour, then you can tell if they still have the "us vs. them" mentality by sandbagging the test and all of them ending up using all the time. Some will work faster than others, and you should be able to match up with their experience and speed. This was one of the techniques I used to know the strengths and weaknesses of not only each drafter, but as a group. Sure, some things take up a lot of time, and some times it's the computers that hold up the drafters..
IMO (personal), if the computer is sitting idle and it's not processing anything or the mouse and commands aren't inputting info, then you're too slow. The mentality is to have everyone on board to get these projects done and out the door so that no one has to stay late or punch in over time (I have better things to do after 5pm). The more ppl do not know where or how fast they are, the more you can sit front of the computer and try to figure that out after 5pm. Its the drafter's choice of not being ahead of the game enough to estimate their time and still retain their job. Purposely sandbagging to be slow is not going to get them a raise either, it becomes a downward spiral of hating their job and being underpaid doing so because the company didnt budget the time it would take to finish it and there's goes all the profit that could have been made.