Overview
Overton analyses millions of policy documents to identify when they cite or mention research from academic and research institutions. To do this accurately, we need comprehensive, up-to-date information about research organisations worldwide.
What is ROR?
ROR (Research Organization Registry) is an open, community-led registry that provides unique identifiers for research organizations globally. It’s maintained collaboratively by a consortium of organizations committed to keeping research infrastructure open and sustainable.
ROR includes:
- Universities and research institutes
- Government research agencies
- Healthcare and medical research organizations
- Related entities like departments, centers, and subsidiaries
Why we migrated from GRID to ROR
Previously, Overton used GRID (Global Research Identifier Database) to identify research institutions. However, GRID stopped receiving updates in Q4 2021, which meant:
- Institution names were increasingly out of date
- New organisations weren’t being added
- Organisational changes (mergers, renamings, restructures) weren’t reflected
- Related organisations were harder to track accurately
In 2026, we migrated to ROR because:
✓ Actively maintained – ROR is continuously updated with new institutions and changes
✓ More comprehensive – Better coverage of institutional relationships and hierarchies
✓ Industry standard – Increasingly adopted by funders, publishers, and research systems
✓ Community-led – Open, transparent governance ensures long-term sustainability
✓ Better data quality – Reduces the need for manual corrections and data cleanup
What changed for users?
New filters
As part of this change we updated some of the filters that were available as part of our document searching. We now allow users to filter documents by whether they:
- Cite a document published from a researcher that is affiliated with the research organisation
- Mention a researcher that is affiliated with the research organisation
- Cite or mention a work or research from a research organisation.
This gives our users more control over how they filter Overton’s data to better understand where they are having impact within policy documents.
Improved accuracy
You’ll now see:
- Current institution names (not historical names from 2021)
- Better detection of citations from related organisations
- More comprehensive coverage of institutional affiliations
Saved searches
Existing saved searches using the old filters will continue to function. However, we recommend updating them to the new filters for optimal performance:
- Navigate to your Saved Searches
- Look for any using the old “Cites or mentions institution”
- Update these to use “Citing research institutions” or “Mentioning research institutions” or “Cites or mentions research institution”
Understanding data discrepancies
When comparing results from before and after the migration, you may notice differences. Here are the most common reasons:
Institution name changes: Organisations may have changed their official name since 2021. Results now appear under the current name.
Better organisational relationship tracking: ROR provides more detailed information about institutional hierarchies, allowing us to better capture citations from departments, centers, and related entities.
Improved affiliation matching: ROR’s comprehensive data allows us to more accurately match institutional affiliations in research papers to policy citations.
Updated organisational structures: Mergers, restructures, and organisational changes since 2021 are now reflected in the data.
In most cases, you’ll see MORE citations
The majority of users will see an increase in citations when comparing searches before and after the migration. This isn’t because we’ve changed how we analyse documents—it’s because we can now more accurately detect and attribute institutional affiliations.
Still seeing issues?
If you notice specific problems with institution data—such as missing organisations, incorrect attributions, or other data quality issues—please let us know. While ROR significantly reduces these issues, we can still manually address specific cases when needed. Contact support.