I'm a child of the Cold War era, when the United States and the Soviet Union were two world super powers locked in an on-going political struggle across the globe. As a child watching the Olympics, it became a running joke to watch politics creep into the judges scores in competition like gymnastics and ice skating. No matter how well an athlete from the U.S. performed, we felt that the scores from the Soviet judges and their allies were always slightly lower than they should have been. Perhaps the Soviet children watching felt the same thing happening from Western judges.
Much like a gymnastics judge watching a routine and trying to place an objective measurement on what took place, call monitoring and quality assessment attempt to objectively measure what took place in an interaction between the company and the customer. There are ways to largely drive subjectivity out of the process and make the assessment as objective as possible. In the end, however, it is a human process. Mistakes will be made. People will see the same thing differently.
This doesn't mean, however, that you throw the baby out with the bath water. I will sometimes hear people argue that QA is completely unreliable because mistakes sometimes occur. If we apply the same logic to athletics than we should just make all the Olympic gymnasts, skaters, and divers hand back their medals and throw the sports out of the competition. That doesn't make much sense, does it?
When done properly, QA can provide a highly accurate picture of the service experience over a period of time. Because it is a human endeavor mistakes will occur. The benefits of listening, assessing, training and coaching far outweigh the risks of sticking your head in the sand and hoping that your team is doing a good job on the phone.