What is DSpace-CRIS? Research information management on top of DSpace

What is DSpace-CRIS for? The problem DSpace alone does not solve

Picture three requests that any research office receives every year:

• "Give me everything Dr. Martinez has produced since she joined the university, with the correct affiliation on each publication."
• "Give me every output of the project funded under call X, with its authors and dates."
• "Produce the institutional CV of this research group for accreditation."

A standard DSpace cannot answer them reliably. It can search for the string "Martinez, Maria" in the author field, but that string coexists with "Martinez, M.", "M. Martinez Rojas" and a version with a married name. DSpace does not know that all four are the same person, or which faculty she belonged to when she published each item. For the funded project there is, at best, free text in a notes field. And the institutional CV is assembled by hand, in a spreadsheet, every year.

The underlying diagnosis is that in a classic repository the person and the project are attributes of the document. In a CRIS they are objects in the system: they exist even without a publication, they have a history, they carry external identifiers and they can be related to each other in both directions. That is why a CRIS can traverse the network of person, unit, project, funder and output, and a repository cannot. In North America this category is usually called a research information management (RIM) system; in Europe, CRIS is the more common term.

What DSpace-CRIS models: the entities

The data model includes at least five types of interconnected entity:

• Researcher profile: the person's record, with name variants, ORCID iD, affiliation history and linked output. It is the entity that solves ambiguous identity.
• Publication or output: the familiar item, now linked to real people instead of text strings.
• Project: with its funding call, period, principal investigator, team and funder.
• Organizational unit: faculty, department, institute or research group, with its hierarchy and its ROR identifier where applicable.
• Funder or grant: the agency and the specific call, so the institution can answer what each grant produced.

The model is configurable: institutions can define additional entities, such as patents, equipment or events, and their attributes. On top of those entities, DSpace-CRIS builds what a repository cannot offer: collaboration network graphs, bibliometrics and statistics aggregated by researcher and department, and CVs and publication lists generated from existing data, without retyping anything.

One technical nuance matters for any project starting today. DSpace-CRIS has historically managed these links through its authority framework, while DSpace 7 and later introduced a relationships framework with a dedicated table. The two models are converging, and DSpace 10 already asks each site to choose which one to use, as explained below.

DSpace vs. DSpace-CRIS, feature by feature

In this comparison, "manual work" means the function can be approximated in standard DSpace, but the system does not deliver it.

The integrations that make DSpace-CRIS valuable

A CRIS without external identifiers is just another database that someone has to fill by hand. What makes DSpace-CRIS useful is that its entities are anchored to the global identifier infrastructure:

• ORCID for researcher identity. The researcher controls the identifier and keeps it when changing institutions. The integration works both ways: researchers can sign in with ORCID and claim their profile, and the system can push the publications and projects linked to the profile to ORCID. An important design detail: whatever the researcher entered manually in ORCID is never overwritten by the repository.
• ROR for organizational identity. The Research Organization Registry is an open registry governed by California Digital Library, Crossref and DataCite, with more than 100,000 organizations and open data. It ensures the university's affiliation is written one way and recognized by Crossref, DataCite, ORCID and OpenAlex.
• OpenAlex as an open index of works, authors and institutions, useful to enrich declared output and cross-check it against what the world already has on record.
• Scopus and Web of Science to import bibliographic records and retrieve citation counts. They require an institutional subscription, but the connection is supported. Open sources such as Crossref, PubMed, SciELO and arXiv are supported too, at no cost.
• DataCite to assign DOIs to datasets and other outputs that do not go through a publisher. It is what turns a deposited file into a citable object.

On reporting, precision matters. Most countries have their own channels for research reporting, such as national research assessment exercises, science agency platforms, funder portals and accreditation dossiers, and in many of them the researcher or the group leader submits the record. A CRIS does not replace those platforms. What it provides is complete, clean data with its supporting evidence in one place, so the institution can feed them without rebuilding the information for every call. In our projects, that is the gain research offices notice first.

Who needs DSpace-CRIS, and who does not

Selling a CRIS to an institution that does not need one is selling complexity, so it is worth being direct.

You need it if your university has active research and reports to a national agency or to funders, if you are going through institutional or program accreditation that requires evidence of output by faculty member and group, if you manage externally funded projects that must be reported per call, if leadership asks for output indicators by faculty, or if researchers lose time rebuilding their CVs every year.

You do not need it yet if the institution is mainly teaching-focused and publishes mostly theses and dissertations, if the repository exists mainly for open access and visibility, or if the repository is not yet populated. That last case matters most: a CRIS on top of an empty repository solves nothing. Fill the repository first, then model the research. When a university arrives in that situation, BiblioLabTech proposes a well-implemented standard DSpace, with the CRIS expansion as a second phase with its own budget.

A real case: NexoHub at UNAB in El Salvador

This is not theory. For Universidad Dr. Andrés Bello (UNAB) in El Salvador, BiblioLabTech implemented NexoHub, an institutional research management system built on DSpace-CRIS. The university's problem was exactly the one described here: it needed to track researchers, units, projects, publications and impact as connected entities, not as loose documents, and it needed unambiguous identity so that the same researcher or unit would not appear duplicated or in conflict across records.

The implementation integrates ORCID for researcher identity, ROR for organizational identity, OpenAlex for bibliographic and citation data and DataCite for persistent identifiers, plus MetaExtract API for automatic metadata extraction in the deposit workflow, Dublin Core metadata, OAI-PMH exposure and managed Linux infrastructure with backups and monitoring. It shows the model works at a mid-sized university, not only at large research-intensive institutions. Universidad Americana in Paraguay, whose DSpace repository BiblioLabTech hosts (read the case study), is evaluating the same expansion toward DSpace-CRIS.

The DSpace and DSpace-CRIS merger: what changes in 2026 and 2027

One development changes how a project of this kind should be planned today. In November 2025, the DSpace Steering Group and the DSpace Leadership Group approved merging DSpace and DSpace-CRIS into a single platform, and the merger is being delivered across two major releases.

• DSpace 10.0, released on May 28, 2026, is the first merged release. It brings into the core a first set of features that used to be exclusive to DSpace-CRIS, including editing items through submission forms, metadata-level security (so researchers can decide, for example, whether the email address or birth date in their profile is public), nested metadata that groups an author with an affiliation, and authority framework enhancements that allow an optional early migration of DSpace-CRIS entities. Each site now chooses between traditional relationships with virtual metadata and DSpace-CRIS-style relationships managed through the authority framework.
• DSpace 11.0, scheduled for May or June 2027, is planned to complete the process and consolidate a single model for relationships between entities, with a migration path from the two current models.

The practical consequences are two. First, CRIS functions stop being a fork of the product and become configurable inside DSpace, so an institution can behave as a classic repository, as a full CRIS or as something in between. Second, and more important for anyone deciding now: starting with standard DSpace does not close the door to a CRIS; each release opens it wider. What pays off is making the modeling decisions from the start, such as normalizing authors, using ORCID and recording affiliations with ROR, because those are what make the later expansion cheap. Our overview of what DSpace is puts the merger in the context of the project's full version history.

How a DSpace-CRIS implementation works in practice

A DSpace-CRIS project looks less like a software installation and more like an institutional data project. These are the phases we use:

1. Define which entities the institution really needs and the attributes of each one.
2. Normalize the list of researchers and connect it to ORCID.
3. Confirm the university's ROR record, and request one if it is missing.
4. Load the history of projects and funding, which usually lives in the research office's spreadsheets.
5. Link the repository's existing output to person and project entities.
6. Only then switch on dashboards, metrics and CV generation.

The step institutions underestimate most is linking the historical output, because there is no button that converts old items into related entities in bulk. It is cleanup and decision work, such as choosing which name variant is correct or which unit a person belonged to in 2019, and it requires someone from the university, not just the provider.

If your repository already runs on DSpace and the institutional conversation has started to turn toward research indicators, now is the time to raise it. If you are still defining the repository, read our DSpace migration and upgrade guide to avoid modeling decisions you would later have to undo, and our breakdown of DSpace implementation cost to size both phases. BiblioLabTech implements both modes, repository and CRIS, through its DSpace and DSpace-CRIS service, and quotes the CRIS expansion separately so that institutions can stage it.