Since 95% is acceptable quality, 98 and 99 are really wonderful.
That is a measure of the PI failures. And we all would like to have such good results. The same media in my drive pulls 97% and the playback quality is excellent.
The percentage doesn't measure the minor PI errors.
Now, you've got to check the minor PI errors yourself to make sure that none of them exceed 350 (top graph on cdspeed). Playability drops above the 100 mark, but that depends entirely on the player in use. For instance Apex players tolerate thousands of errors while Phillips tolerate nothing over 50.
The scan does not check the data integrity (unless it specifically mentions PO failure). It does not directly address the issue of data integrity. It is for a measure of "playability." Of course a better scan makes for a better viewing experience and a longer lasting disc (given that disks gather errors over time, so a well written disc has less total errors now, and later).
Some media, like CMC has beautiful scans and great playback. However the reason that movies on CMC media get stuck sometimes is because there are intermittent data integrity failures that do not show up on scans. As you can see, both tests are useful. However an actual copy or "play" test is much more important than any scan.
To test data integrity, disregard verify functions of recording software. Next, copy the files back to the hard drive using rip software or windows explorer. Or, use the scandisc feature; but, that is slower and less effective than the copy method. With either method, if the process completes without error, the disc has good data integrity regardless of the QC scan.
In fact, I just burned several Ricoh 01 discs that Q-scan said not to use under any speed. Of course I burned these at max and the resulting quality scan was 97%. Data integrity was 100%. Playback was 100%.
Then I burned a Daxon AZ2 (Daxon=BenQ AZ2=cheapskate version). The scan was horrifying (98% due to low PI failure; BUT, minor PI errors were 250). Data integrity was 100%. Playability in a set top player was moderate entertainment value (not 100%). While it did not get stuck (due to 100% data integrity), it did view with many events of jerky motion (as the bad scan predicted).
I hope that comparison helps!
So if your rather perfect scans will copy back to the computer with zero errors and then play in a set top player with mostly smooth motion, I would say that things are going very well for you.