Using adaptive, personalized courses is rewarding, as it can create a better learning experience, tailored for a specific learner?s needs. The process of creating these courses, however, is cumbersome and time consuming. Authors can be helped by employing automatization methods that relieve the authoring burden. We previously proposed a method which uses information from an author?s Desktop to suggest extensions to existing courses. In this paper we report an experimental evaluation of the method, focusing on usability aspects.