You need to use a basket of roughly ten to twenty recordings to test your equipment, tweaks, room, cleaning regimen and the like.
Don’t rely on any given recording to be The Truth. None of them are.
To illustrate this idea, imagine your stereo as a huge diamond. Each and every recording you play is showing you a different facet of that diamond, a different characteristic of your system’s reproduction.
Audiophile X will play a record and say it has bad bass. His bass is excellent on other recordings, so this record, which seems to have bad bass, must be at fault.
Audiophile Y plays the same record and says it has good bass. Assuming the record has good bass for a moment, what in fact is happening in Audiophile X’s system is that most facets of his bass are good, but some facet of his bass is bad, and this record is showing him some shortcoming in his bass reproduction that his other records aren’t showing him.
If Audiophile X makes some changes to his stereo, and the record in question now has better bass, and, importantly, other records still sound as good or better than they used to, then some measure of success will have been achieved, and another step forward will have been taken in that very long and often frustrating journey we are all on.
The diamond has many flaws. We find them and fix them, one at a time.
Along those lines, there is a section of this blog called Difficult to Reproduce Recordings. Many of these albums make great test discs once you have determined through a shootout that you have a top quality copy to work with.