Institutional Research Visibility
How publishers deposit metadata for your institution's journal articles — observed directly from Crossref work records. Scope and window are chosen to respect the free rate limits offered by Crossref and OpenAlex.
How this analysis works
This page measures what publishers deposit to Crossref for your institution's journal articles. Crossref stores what it receives from publishers — the deposit practices are the variable. The analysis uses three data sources together:
1. OpenAlex — for article discovery
The tool queries OpenAlex for journal articles where any author is affiliated with this institution's ROR in the last 90 days. OpenAlex aggregates affiliation signals from multiple sources (Crossref, PubMed, MAG, author pages), so it reliably identifies "these are the papers that came out of your institution" even when one source is incomplete.
2. Crossref — for observed publisher deposits
For every DOI OpenAlex returns at each mapped publisher, the tool fetches the actual Crossref work record and inspects it directly. Each record is counted by what it contains — affiliations, ROR IDs, funders, abstracts, licenses, ORCIDs — exactly as the publisher deposited them.
3. ROR — the interlingua
ROR IDs are the machine-readable bridge. OpenAlex asserts "this paper is from institution X"; the publisher's Crossref deposit may or may not include X's ROR on any author. The Inst. ROR column tests whether the publisher included this institution's specific ROR in the deposit — the gold-standard signal for automated institutional linking.
How the numbers are built
- Gap counts are observed from each Crossref work record, not projected from publisher-wide averages.
- If OpenAlex identifies a paper as institutional output but the Crossref deposit has no institutional ROR, that's counted as a gap — the affiliation exists, the publisher just didn't deposit it.
- Only journal articles are inspected — no proceedings, book chapters, or peer reviews — to avoid the content-type dilution that inflates aggregate-style scores.
- Publishers with fewer than 10 articles from this institution are shown but not measured (sample too small).
- Articles that can't be traced to a mapped Crossref deposit are reported separately in "Why N articles weren't analyzed" with a per-reason breakdown.
- A single extra OpenAlex count call fetches the institution's full-year journal-article output; the cost calculator uses that number to extrapolate the observed 90-day gap to an annual rate at your chosen hourly rates.
What to keep in mind while reading the report
- Not every article can be analyzed — and every reason is a deposit gap. Some articles were never registered with Crossref (published only in an institutional repo, preprint server, or data platform); some come from publishers not yet mapped to a Crossref member ID; some have a deposit so incomplete that the publisher itself can't be identified downstream. The "Why N articles weren't analyzed" section below breaks these down.
- Author attribution is permissive. A paper is treated as "from this institution" if any author is affiliated with it. A co-authored paper with 20 authors across 5 institutions counts once for each — the deposit may or may not actually include this institution's ROR even when the institutional link is real.
- String affiliation without ROR is not "missing". A publisher that deposits "University of Oxford, Department of X" as text but no ROR ID will show as has affiliation but no ROR. The attribution exists in human-readable form; machine processing is what's blocked.
- ROR has no funder-vs-institution type flag. A single organization (Wellcome, Max Planck, NIH, etc.) holds one ROR ID that can play either role — the meaning depends on where the ROR appears in the deposit: in
author.affiliation.idit's an institutional signal; infunder.idit's a funder signal. This tool currently inspects ROR only in the affiliation field. As publishers transition from the legacy Open Funder Registry to funder ROR IDs, a separate funder-ROR metric may be added. - Point-in-time snapshot. Publishers can and do revise deposits. Numbers here reflect the state of Crossref at the moment of analysis — don't treat them as immutable.
- Small samples are omitted. Publishers with fewer than 10 articles from this institution are shown but not measured — too few records to draw meaningful percentages from. Counts below that threshold are for reference only.
- Annual cost estimates are extrapolated. The hours shown per stakeholder come from the observed 90-day gap, scaled up to a year using the institution's real full-year article count. Hourly rates you enter in the cost calculator stay local to your browser.
What each gap column means
- No Affiliation — the Crossref deposit carries no author affiliation at all. Not just "missing ROR" — the affiliation text field itself is empty. Institutional linking from the deposit is impossible without manual publisher-site scraping.
- No Any ROR — the deposit has affiliation text but no ROR identifier on any author. Matching still requires string parsing of "Department of X, University of Y" — brittle and fails for renamed departments, translated names, and multi-institution affiliations.
- No Inst. ROR — the deposit does not include this specific institution's ROR on any author. This is the strictest test. A paper may have ROR IDs for other co-author institutions but not this one's. CRIS auto-ingest requires this exact match.
- No Funder — the deposit has no funder metadata. The paper may have been funded; the funder-grant information simply wasn't transmitted. Mandate-compliance reporting (UKRI, NIH, Wellcome) falls back to manual grant-database cross-referencing.
- No Abstract — the deposit has no abstract. Discovery tools (Semantic Scholar, OpenAlex, Elicit, Consensus) index from Crossref abstracts. Missing = less discoverable, fewer citations, weaker research impact signal.
- No ORCID / License — similar story for author identity persistence (ORCID) and open-access verification (license). Both are schema-supported; both are deposit-practice dependent.