Current Research Information Systems (CRIS) have been a critical institutional infrastructure layer for two decades, capturing researchers, their outputs, their funding, their collaborations, their projects. The three dominant data models in 2026 are CERIF (the European standard model, maintained by euroCRIS), Pure (the Elsevier-operated CRIS, with the largest market share in research-intensive universities), and VIVO (the open-source community-maintained CRIS with strong North American adoption). The three are convergent in intent and divergent in detail. This post is a practical guide to interoperating across them.
What each model is
CERIF is the Common European Research Information Format, maintained by euroCRIS since 2002. CERIF is a data model, not a system; it specifies the entities and relationships a CRIS should track and how they should be expressed in XML or RDF. CERIF-CRIS systems exist in many forms (the original CERIF reference implementations, Elsevier’s Pure with CERIF compliance, various national-system implementations) and CERIF compatibility is the lingua-franca claim in the European CRIS market.
Pure is the dominant commercial CRIS product, used by hundreds of research-intensive universities globally. Pure has its own data model, which is broadly CERIF-compatible but with vendor-specific extensions and refinements. Pure’s market position means its data model functions as a de facto standard regardless of its formal status.
VIVO is the open-source community-maintained CRIS originally developed at Cornell and now maintained by an international community under the DuraSpace umbrella. VIVO is built on a semantic-web foundation with an RDF/OWL ontology and explicit federation-friendly design. VIVO has strong adoption in US research universities and a growing international community.
Where the models align
The three converge on the core entities. All three model researchers (with ORCID iDs), organisations (with ROR IDs), publications (with DOIs), projects (with funding metadata), and the relationships between them. The CRediT roles can be expressed in all three. The funder-grant-output structure is representable in all three. For the 80% of routine queries against a CRIS, all three produce comparable answers.
The convergence has been substantially driven by external standards. ORCID, ROR, DOIs, Crossref Funder Registry, CRediT, RDA DMP Common Standard — these external persistent-identifier and metadata standards have pulled the CRIS models toward common representations even where the internal models differ. The CASRAI research information systems domain tracks the convergence.
Where the models diverge
Three areas of substantial divergence.
First, granularity of activities. CERIF models a wide range of research activities at different granularities (projects, work packages, deliverables, milestones); Pure focuses on the publication-centric workflow with project as a supporting entity; VIVO’s ontology accommodates both but is community-extended in ways that vary by deployment. An institution moving CRIS from one platform to another typically loses or transforms activity-level data in ways that require careful migration planning.
Second, contribution and contributorship. CERIF’s contributor structure has evolved to carry CRediT roles natively. Pure carries CRediT but with vendor-specific extensions. VIVO’s ontology can express CRediT but the per-deployment representation varies. A research output with structured contributorship in one CRIS may lose detail when exported to another.
Third, extension and customisation. Pure customers heavily customise their deployments with institution-specific fields and workflows; VIVO sites likewise extend the ontology. The customisations are valuable locally and problematic for cross-institutional interoperability. A federated query that works at one institution may return different fields at another, even where both claim CERIF compliance.
The interoperability layer
The practical interoperability layer in 2026 runs through three exchange mechanisms.
OpenAIRE-CRIS is the European interoperability profile for CERIF, defining a subset of CERIF that all participating CRIS systems can emit and consume. OpenAIRE consumes CERIF-CRIS feeds via OpenAIRE-CRIS and incorporates them into the OpenAIRE Graph. Most European institutional CRIS systems can produce OpenAIRE-CRIS-compliant feeds with modest configuration.
ORCID-CRIS integration is the per-researcher exchange channel. A CRIS depositing publication and affiliation data to ORCID, and consuming corrections back from ORCID, becomes a node in the ORCID-anchored researcher record. All three major CRIS models support ORCID integration, though the depth varies.
Crossref event data and citation feeds provide the publication-level exchange. A CRIS that ingests Crossref event data picks up post-publication corrections, citations, and relationship updates that the local CRIS would otherwise miss.
The three exchange mechanisms together cover most of what cross-institutional interoperability requires. They do not cover the activity-level data that diverges across CRIS models; that data remains harder to interoperate.
What institutions should do
For institutions selecting or migrating a CRIS, the practical recommendations are: prioritise CERIF compliance regardless of vendor; require ORCID integration; require Crossref event-data ingestion; verify OpenAIRE-CRIS compliance for institutions with European funder reporting obligations; insist on data-export capability that includes the full activity-level data, not just the publication-centric subset.
For institutions operating an established CRIS, the priorities are to keep the integration layers current (ORCID 4.0 transition, OpenAIRE-CRIS profile updates, Crossref REST API consumption), to invest in metadata-quality QA, and to participate in the CERIF, Pure, or VIVO community work to influence the data-model evolution.
For CRIS vendors, the priorities are to honour the convergent standards (ORCID, ROR, CRediT, OpenAIRE-CRIS) without burying them under vendor-specific extensions, and to make data export and import paths reliable across customer transitions. The market would benefit from less lock-in friction; the standards work supports that direction.
The euroCRIS-DuraSpace-Elsevier triangle
Beneath the technical layer is an organisational layer. euroCRIS as a standards body, DuraSpace as the VIVO open-source community steward, Elsevier as the Pure operator — these three together substantially set the direction of CRIS evolution. The 2024-2025 coordination work (visible in the joint CERIF-VIVO ontology alignment, the Pure CERIF-compliance certification process, the OpenAIRE-CRIS profile refinement) has been more productive than the prior decade.
The convergence is incomplete and uneven, but the direction is clear. By 2028, cross-CRIS interoperability for the standard entities (researchers, outputs, projects, funding) should be a routine technical exercise, not a multi-year integration project. The activity-level interoperability will follow more slowly.
