At its inception in 1989, the DFSMS Data Class construct consisted of a simple list of attributes that could be used to describe the physical characteristics of a dataset. Now, more than three decades later, the Data Class includes more than 50 different values that can interact in complex ways with other SMS constructs such as Storage Class and with JCL and IDCAMS control statements. Some user SMS configurations are home to dozens or even more overlapping or conflicting data classes. This webinar aims to provide an understanding of what the important (and unimportant) data class values are, their history, and their relationship to z/OS dataset allocation and use.