Robotic Process Automation (RPA) has gained much attention both in industry and academia. One of the main challenges for a successful implementation of RPA is selecting which tasks should be automated. While different methods exist to identify RPA candidate tasks, they lack in providing objective evidence on why to automate that task. Such objective evidence can be gathered by applying process mining techniques to gain insights into the performance of a process and its tasks. Although this has multiple advantages, it can be time-consuming to analyze all potential processes. We conducted a literature review of existing methods to identify relevant criteria and method components, based on which we designed a framework for identifying and prioritizing suitable RPA candidate tasks: the Prioritized List of Suitable Tasks (PLOST) Framework. The framework includes both qualitative and quantitative assessment criteria and guides the analyst to focus on relevant processes before zooming in on the task level. It also takes into account a customized automation strategy. We conducted a case study to evaluate the applicability and effectiveness of the PLOST Framework and performed thinking-aloud sessions to evaluate its usability, practicality, and completeness. The results show that the framework is easy to apply and feasible.