We agree with the common perception that Mobility as a Service (MaaS) has strong potential to address the growing mobility needs of the world's population in a sustainable way. However there is a significant lack of understanding of what MaaS actually is on conceptual level. Reviewing the state-of-the-art of MaaS research we deduct that MaaS has all the attributes of a ‘hyped’ socio-technical phenomenon: it seems to be a loosely connected patchwork of optimistic political dogma, activists' enthusiasm, anecdotal evidence of successful services and a firm belief of investors in companies such as Uber. However, as a paradigm shift, MaaS has electrified public decision makers and unleashed a wave of innovation in SMEs. In addition, MaaS forces researchers to work in multi-disciplinary collaborations, transcending ‘traditional’ competence areas, such as traffic planning, computer science, social sciences, transport systems and organisational psychology. Consequently, while embracing the innovative spirit of the MaaS paradigm shift, we aim to outline key conceptual issues of MaaS. These issues are 1) the nature of travel (including purpose, mode, means, distance etc.); 2) MaaS interoperability within intelligent traffic services; 3) the end-user perspective, with a focus on user attitudes and behaviours; and 4) sustainability, including work during the journey, and ultimately not traveling at all (e.g. teleworking). We exemplify our tentative concept through applying it to the existing mobility services Uber, Lyft, Car2Go, DriveNow, Tuup, UbiGo and Moovel. Through our study, we provide an initial research based conceptual foundation for MaaS researchers, developers and implementers.