This poster presents a specific approach how to establish data stewards for health-related sciences at UAS which is currently being developed as part of a research project. Particular emphasis is placed on skills that are essential in order to provide subject-specific RDM-support, taking into account the reality of limited resources. As the implementation process in our research project is well underway, we also share our experiences of introducing subject-specific data stewards, including hurdles and lessons learned.The idea of the presented concept is to implement local subject-specific RDM-support at selected UAS aiming to establish RDM at an institutional level. To achieve this, one data steward per UAS (0,5 full-time equivalent each) has been employed. The data stewards are employed on a temporary basis for three years corresponding to the duration of the project. Within the scope of this project, our approach is currently being tested at two German UAS, which previously had no RDM staff.Overall, change is slow and patience is needed to establish RDM at an institutional level and, most importantly, in the minds and practices of researchers. Clearly, need for RDM supports is substantial and will persist long-term. As a result, most of the work that data stewards undertake is of a permanent nature. This raises the question of how services and processes that are currently being established can be maintained and made sustainable permanently. The approach of establishing local data stewards at UAS has some advantages, as researchers need a local contact person that they are familiar with and can reach easily. At the same time, local data stewards are able to gain deep insights into researchers’ RDM needs. Establishing at least one data steward per institution, as recommended in this approach, can be seen as a first important step towards institutionalising RDM at UAS.