Detailed observation of user behavior is indispensable for software companies such as SAP. Only in this way can it be determined whether the actual behavior corresponds to the intended use. If this is not the case this canhave a negative impact on the efficiency of the executed business processes and the perceived usability of the software. “User behavior mining” tracks how users interact with software to process business objects. For example, it is possible to observe which steps were taken to process customer orders and invoices. The overall vision of this research project by Prof. Dr. Jana Rehse in collaboration with SAP Business Process Intelligence (BPI) is to improve user support and experience and to uncover situational automation potential within SAP systems. To this end, existing methods from process mining are being adapted and further developed for application to UBM-data. For example, a cluster analysis approach was applied to identify different user groups in the data. This makes it easier to distinguish between the needs of individual users and to adapt the software accordingly. Furthermore, an approach for predicting user behavior based on machine learning was implemented. In future, this will enable users to navigate more quickly within the software by means of a corresponding assistance function. Due to the amount and high resolution of the data, a purely data-driven approach is possible.