Personally I utilize a couple different models and compare the results … hopefully if things are going as one would expect the percentages should be relatively close.
I understand that some individuals utilize casino popups – This undeniably inflates their hit counts and thus for these webmasters to analyze hits to reals wouldn’t be a very useful. I do however think this is the most useful percentage though as you would be able to figure a more precise ROI as Simoneaton alluded to.
Personally I don’t see how popping a casino in the background could be that beneficial – heck more than likely you’ve irritated the hell out of your visitor(s.) That would be another conversion percentage that would need to be assessed via tracking different campaigns.
I believe that some programs may utilize click quotas; that is to say that X amount of traffic is worth a range of earnings Y amount of traffic is worth another range, Z … and so forth – you get the picture. Unless you are sending a specific amount of traffic you will not qualify for anything but a healthy shave.
The objective of this exercise should be to determine an average, a mean, and a median percentage for each casino and thus identify any patterns that may exist among a broad range of affiliates and programs. I don’t think providing a % would be giving away too much info. To say traffic to program A converts at 10% and traffic to program B converts @ 21% doesn’t identify the amount of traffic.
Maybe this is something the APCW can/will/or already is attempting to address.