The 80-20 rule is certainly alive and well with keyword advertising. It's true that 20% of your keywords will drive 80% of your volume. The question comes down to identifying the 20% as fast and as efficently as possible.

I have worked with some clients who's search logs indicate the terms people have found them under, and this has often been a great place to start. However this is often only part of the story. If they don't have the specific terms on their site, they haven't been found under them, right?

I agree with Todd, 100% when he says you need to be able to run quick tests to see if a program is going to work before investing the time on extensive keyword research.

Once you have found that great program though, I have been very sucessful in testing large keyword lists as a way to identify the 20% that will drive my volume.

It's getting a lot harder though to run any massive lists for a program, I might have to look at more of a staging approach, launching a large list of quality terms and letting them come off hold before launching the more extensive lists that I have been able to build.

I don't know which approach is best, and certanily I have used both approaches at different times where appropriate. I am sure Todd has as well.

The bottom line is it's getting harder to work with Google in this fashion and that means it's time to adjust to the changing times or advertise elsewhere...