In this paper we describe the development of a planning tool for a department of commercial credit applications of a large player in the Belgian banking sector. The tool is created to cope with service level agreements (SLA) that put a limit on acceptable throughput times of credit requests. Currently, SLA requirements are not well met and customer satisfaction is low. Because of the highly variable process times and complex flows with rework loops, a deterministic scheduler is not appropriate. Simulation is therefore used to model the processes. Firstly, matching simulation results with real data reveals if the processes are lean, and fine-tuning can be done. Secondly, based on simulation results a tool can be developed representing expected throughput times and SLA impact for different degrees of capacity utilization. Consequently, different workforce allocations can be tested and deliberate choices can be made.