Design science research (DSR) is an appropriate method for postgraduate-level (MSc) thesis work, where the students need to combine and apply theoretical knowledge and hands-on skills to solve practical problems. Students face considerable challenges, however, when applying DSR in timeboxed situations with limited competences. In this paper, I reflect on my experiences supervising more than 200 MSc theses, many using--and some that intended to use--DSR. My reflections provide teaching tips to educators and supervisors and show a need for new DSR process and evaluation methods.