In this article, we present the prototype of a modeling tool that applies graph-based rules for identifying problems in business process models. The advantage of our approach is twofold. First, it is unnecessary to compute the complete state space of the model to find errors. Second, our technique can even be applied to incomplete business process models. Thus, the modeler can be supported by direct feedback during the model construction. This feedback not only reports problems, but also identifies their reasons and gives suggestions for improvement. Copyright © 2010 John Wiley & Sons, Ltd.