PurposeThe purpose of this study is to explore current practices, challenges and technological needs of different data repositories.Design/methodology/approachAn online survey was designed for data repository managers, and contact information from the re3data, a data repository registry, was collected to disseminate the survey.FindingsIn total, 189 responses were received, including 47% discipline specific and 34% institutional data repositories. A total of 71% of the repositories reporting their software used bespoke technical frameworks, with DSpace, EPrint and Dataverse being commonly used by institutional repositories. Of repository managers, 32% reported tracking secondary data reuse while 50% would like to. Among data reuse metrics, citation counts were considered extremely important by the majority, followed by links to the data from other websites and download counts. Despite their perceived usefulness, repository managers struggle to track dataset citations. Most repository managers support dataset and metadata quality checks via librarians, subject specialists or information professionals. A lack of engagement from users and a lack of human resources are the top two challenges, and outreach is the most common motivator mentioned by repositories across all groups. Ensuring findable, accessible, interoperable and reusable (FAIR) data (49%), providing user support for research (36%) and developing best practices (29%) are the top three priorities for repository managers. The main recommendations for future repository systems are as follows: integration and interoperability between data and systems (30%), better research data management (RDM) tools (19%), tools that allow computation without downloading datasets (16%) and automated systems (16%).Originality/valueThis study identifies the current challenges and needs for improving data repository functionalities and user experiences.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/OIR-04-2021-0204