Metadata completeness — one score
Any journal by its ISSN, any DOI prefix, any Crossref member: how much of the metadata indexes read is on the record, what is missing, and who fixes it. Free, no account — every figure is Crossref's own.
What large publishers reach, read on 16 Oct 2026
Each scored exactly as any member is scored here — from its own Crossref member record, by the same seven signals and weights. Names are Crossref's own. The median of the 12 is 41.
How the score is made
- Reference lists — impact 22
- Abstracts — impact 18
- Licences — impact 16
- ORCID iDs — impact 14
- Author affiliations — impact 12
- ROR institution ids — impact 10
- Funder data — impact 8
The score is the share of each signal on the record, weighted by its impact (the impacts sum to 100). A signal Crossref gives no figure for is left out and the rest are weighed as before — never counted as 0. Grade A from 85, B from 70, C from 55, D from 40, otherwise E. A journal is read off Crossref's journal record (its current and backfile DOIs, weighted by how many each holds); a prefix and all of Crossref by counting every DOI with Crossref's own has-… filters; a member off its member record, as What Crossref holds about you reads it. The members door's engagement dashboard prints the same score from the same functions.
Sources: Crossref REST API · Crossref participation reports · ROR · ORCID · doi.org. This page reads only what is public; it says what a database sees when it looks the record up — not whether a journal is well run.
