In enterprise software projects, we can easily experience performance problems while developing a software product. The periodic performance regression testing againt the developed code is very essential for more rapid feedback of performance problems. Detecting and analyzing performance anomalies require much manual work. Also we need the clear criteria to identify the anomalies. In our previous research, we have introduced the framework for performance anomaly detection and analysis, which can detect performance anomalies automatically using statistical process control in the software development process. In this paper, we categorize the types of performance anomalies, which have been found while applying the proposed system on our software product. Also we present how the existing performance anomalies can affect the automated anomaly detection system and how to avoid the effects.