The infrastructure of persistent identifiers has, over two decades, learned to name most of the things that matter in research. A DOI names an output. An ORCID iD names a person. A ROR identifier names an organisation. A grant DOI names an award. Yet one obvious entity went unnamed for a surprisingly long time: the research project or activity itself, the container within which people, grants, organisations and outputs all come together. RAiD, the Research Activity Identifier, was created to fill precisely that gap, and it provides the missing layer that lets the rest of the identifier graph cohere around the work as a whole.
What RAiD identifies
A RAiD is a persistent identifier for a research project, or more generally a research activity. It deliberately occupies a layer distinct from the entities the other identifiers describe. ORCID identifies the who, ROR the where, and DOIs the what that is produced; RAiD identifies the activity, the bounded piece of research work that those people, organisations and outputs are associated with. In practice a RAiD record describes a project: its name, its start and where relevant its end, the people involved and their roles, the organisations participating, the funding awards attached to it, and the outputs it produces. Crucially, a RAiD is intended to evolve over the life of the project, accumulating links as new people join, new funding arrives and new outputs appear, so it functions as a living index of the activity rather than a static stamp.
The value of naming the activity is that it turns a scattered set of separately identified things into a connected whole. Without a project-level identifier, the relationship between a grant, the researchers it funded, the institutions they worked at and the papers and datasets they produced must be reconstructed by inference. With a RAiD, those relationships are asserted explicitly in one place, and any system can resolve the RAiD to see the project and everything connected to it.
An international standard: ISO 23527:2022
RAiD is not merely a useful convention; it is a formally standardised identifier. It is specified in ISO 23527:2022, the international standard for the Research Activity Identifier published by the International Organization for Standardization. Standardisation matters for an identifier intended to be adopted globally and to interoperate with other systems: it means the structure, semantics and governance expectations are defined in a stable, internationally recognised document rather than left to a single provider’s discretion. An ISO standard gives funders, institutions and infrastructure providers the confidence to build on RAiD, knowing it rests on an agreed specification rather than a proprietary one.
The role of the ARDC
RAiD originated from and is championed by the Australian Research Data Commons (ARDC), the Australian national research-infrastructure body. The ARDC developed RAiD to address the project-level gap in the identifier landscape and has been central to its standardisation and to operating it as a service, working with international partners to establish the governance and registration arrangements that allow RAiDs to be minted and maintained. The ARDC’s involvement situates RAiD within the broader national and international research-infrastructure community, alongside the registries that run the other persistent identifiers, rather than as an isolated tool.
How RAiD links the graph together
The point of RAiD is connection, and it earns its place by linking to the established identifiers rather than replacing them. A single RAiD can tie together:
- Data management plans. A project’s DMP, which sets out how its data will be handled, can be associated with the RAiD, anchoring planning documents to the activity they govern.
- Grants. The funding awards supporting the project, themselves identifiable through grant identifiers, can be linked, so the money and the activity it pays for are explicitly connected.
- People, via ORCID. The researchers involved are linked through their ORCID iDs, with their roles in the project recorded, so contribution can be tracked at the activity level.
- Organisations, via ROR. The participating institutions are identified by their ROR identifiers, situating the project within the verified organisation registry.
- Outputs, via DOI. The publications, datasets, software and other outputs the project generates are linked by their DOIs, closing the loop from activity to result.
Assembled this way, a RAiD becomes the hub of a small graph centred on the project, with spokes reaching out to people, organisations, funding, plans and outputs, each named by its own appropriate persistent identifier. This is the architecture the persistent-identifier community has long described as a connected graph of trustworthy metadata, and RAiD supplies the node that was missing from it. It complements rather than competes with the others, and shared definitions of the kind a CASRAI data dictionary provides help ensure that the roles, statuses and relationships recorded against a RAiD mean the same thing across systems.
Why it matters
For funders and institutions, project-level identification promises a far cleaner way to report and analyse research. Reporting on a project, tracing what a grant produced, or understanding how a body of work developed becomes a matter of following links from a single identifier rather than stitching together evidence by hand. For researchers, a RAiD offers a durable home for the project that persists as people, funding and outputs change around it, and that connects their contributions, recordable through frameworks such as CRediT, to the activity in which they were made. As adoption grows, RAiD has the potential to make the entire identifier ecosystem more coherent by giving everything a shared centre of gravity: the research activity itself.
