In the UK and US, many funding bodies require applications for research funding to include a data management plan (DMP). To assist researchers with writing these plans, services such as DMPonline and DMPTool provide a range of online templates that are customised according to requirements of particular funding bodies. While research funding bodies in Australia do not generally require a formal DMP in funding applications, creation of a DMP is, at some academic institutions, a condition for access to research infrastructure such as archival storage for data. A related expectation, from funders and some academic publishers, is that metadata and data associated with publications are documented and made accessible to other researchers, or to industry and community groups. These trends reflect the growing recognition that research data are valuable research outputs and an important component of scientific and scholarly communication.

The paper raises questions about the impact of these drivers for documenting and managing metadata and data on the actual practice of research. Is there a risk that tools and procedures that are designed to help researchers acquire funding, publication or access to research infrastructure will compromise the integrity of metadata or limit possibilities for re-use of data? Publication of data with minimal metadata, in order to fulfil requirements of funders or publishers, may aid discovery of the data without adequately supporting its re-use – or may lead to subsequent research findings based on incomplete or inaccurate contextual metadata.

These risks can be minimised if tools and procedures for managing and publishing data are embedded in research workflows as well as administrative and reporting processes. It is important that infrastructure for data-related planning, managing and reporting interoperates with standards for disciplinary metadata and protocols for research practice.