Wow, that's tough, but doable. Your ETL's just need to make sure they know how to read a monitor correctly. If they are holding you accountable by what the actual monitor is actually saying, then it's not fair. Sometimes people forget to log out of batches (they shouldn't however!), 2 people get into a batch, Or there are some quirky batches like PRO1, BAKE, etc; that sometimes seem to be off. For some reason CNDY, MM2, Softlines, give you even more than half the time to complete the batches.
What I did once, was analyze how long RWT was giving team members to complete certain batches. I realized a better indicator of speed was to analyze the "# of Locations scanned" when the pulls were finished! If someone is doing all receiving batches, while someone else is doing easier ones like softlines and market, the 2nd person is going to look a hell of a lot faster, when that might not be the case at all. So I gave each batch a "degree of difficulty" to make it fair. I gave receiving pulls a little more slack and made it harder on the easy batches. When the pulls were finished, I took the monitor, multiplied the locations scanned by the degree of difficulty and ended up with some nice numbers. I shared the results with my team and they started averaging 45%-50% (real) pull times. Made a contest out of it.