Science journals sit inside a much larger scholarly-record system. Crossref reports 183,269,232 total records, while PubMed comprises more than 40 million citations. Clarivate’s 2024 Journal Citation Reports (JCR) covers more than 21,800 journals, showing why journal statistics need clear definitions: a database record, a citation, a journal, and a DOI are related but different measurements.
Contents
- How large is the scholarly record?
- Journals, proceedings, and publication objects
- DOIs and the connections between records
- Recent Crossref deposit activity
- What the monthly figures show
- Journal coverage and open access
- Metadata, volunteers, and report reach
How large is the scholarly record?
Crossref’s Status Report lists 183,269,232 total records in its database. That total includes several kinds of scholarly and research-related objects rather than journals alone. Crossref’s APIs page describes a metadata corpus of over 165 million research outputs from over 22,000 members worldwide. The difference between these figures reflects different descriptions and reporting contexts; neither should be treated as a count of science articles alone.
PubMed offers another important view of the literature. Its home page says the service comprises more than 40 million citations. A citation database is not interchangeable with a DOI registry: PubMed’s figure describes citations in PubMed, while Crossref’s totals describe deposited metadata and registered records.
Crossref also reports 34,856 total prefixes. A prefix identifies the registrant namespace used in DOI assignment, so it provides context about the organizations contributing records. Crossref’s 2024 public data file contained over 156 million metadata records deposited through the end of April 2024 and came from over 19,000 members. These figures describe a particular public data file and cutoff, not a timeless total.
Journals, proceedings, and publication objects
Crossref’s Status Report lists 163,709 journals and 139,625 conference proceedings. Those two categories alone demonstrate why a broad database total cannot be read as a journal count. The same report lists 70,435 databases, 505,789 standards titles, and 9,317,228 components.
The report also lists 199,944 grant DOIs and 1,038,789 peer review DOIs. These identifiers point to research-related objects that can support transparency and traceability around published work, but they are not additional journal titles.
Here is a compact comparison of selected Crossref record categories:
| Category | Count reported by Crossref |
|---|---|
| Total records | 183,269,232 |
| Journals | 163,709 |
| Conference proceedings | 139,625 |
| Databases | 70,435 |
| Standards titles | 505,789 |
| Components | 9,317,228 |
| Grant DOIs | 199,944 |
| Peer review DOIs | 1,038,789 |
The categories should not be added together. They describe overlapping dimensions of the registry: some values count objects, some count titles, and some count DOI records. Keeping the labels attached is essential when using these statistics in a science guide.
DOIs and the connections between records
Crossref lists 123,504,176 journal DOIs and 9,880,213 conference DOIs. It also lists 3,475,367 database DOIs and 414,267 standards DOIs. These counts concern DOI records associated with the named object types, not necessarily unique publications in every practical sense.
Crossref’s Status Report lists 81,448 journal title-level DOIs, 2,427,687 book title-level DOIs, and 82,571 conference proceeding title-level DOIs. It further lists 1,061,310 dissertation title-level DOIs, 951,927 report title-level DOIs, 413,702 standards title-level DOIs, and 58,199 database title-level DOIs.
FundRef information provides another way to describe metadata coverage. Crossref lists 2,947,660 unique DOIs with FundRef data. This is a subset defined by the presence of that funding metadata, not a count of all funded studies or all grant-related records.
The citation-linking scale is much larger than the title-level DOI counts: Crossref lists 2,005,894,912 cited-by links. A cited-by link represents a relationship between records. It should therefore be read as a network measurement rather than as a count of articles, journals, or unique citations in a separate index.
Recent Crossref deposit activity
Crossref’s annual figures show how much activity can occur in one reporting year. In 2023, the monthly deposit total was 12,069,831 records, including 10,417,836 current deposits and 1,651,995 backfile deposits. The same report lists 49,255,880 total updated records for 2023.
For 2024, the monthly deposit total was 10,886,577 records: 9,424,737 current deposits and 1,461,840 backfile deposits. Total updated records reached 53,544,428. These are annual report figures and should not be interpreted as a direct estimate of the number of newly published science articles.
Crossref’s 2025 figures list 12,137,579 records in the monthly deposit total, made up of 10,302,272 current deposits and 1,835,307 backfile deposits. The report lists 78,775,016 total updated records for 2025. A deposit is a registry event, while an update can affect metadata for a record already present.
| Reporting year | Monthly deposit total | Current deposits | Backfile deposits | Total updated records |
|---|---|---|---|---|
| 2023 | 12,069,831 | 10,417,836 | 1,651,995 | 49,255,880 |
| 2024 | 10,886,577 | 9,424,737 | 1,461,840 | 53,544,428 |
| 2025 | 12,137,579 | 10,302,272 | 1,835,307 | 78,775,016 |
What the monthly figures show
Crossref recorded 999,607 added deposits in April 2025, including 848,400 current and 151,207 backfile deposits. It recorded 8,476,206 updated records that month. In May, added deposits totaled 980,672, with 846,833 current and 133,839 backfile deposits; updated records totaled 4,865,373.
June 2025 brought 964,716 added deposits: 834,582 current and 130,134 backfile. Crossref recorded 4,398,874 updated records. In July, added deposits reached 991,670, consisting of 891,425 current and 100,245 backfile deposits, while updated records reached 6,437,528.
August recorded 1,049,095 added deposits, including 826,413 current and 222,682 backfile deposits. Updated records totaled 6,287,301. September recorded 940,590 added deposits, with 845,229 current and 95,361 backfile deposits, plus 6,644,924 updated records.
The final quarter in the supplied Crossref report also shows variation. October recorded 1,052,464 added deposits, including 957,501 current and 94,963 backfile deposits, and 8,869,911 updated records. November recorded 1,022,102 added deposits, including 909,194 current and 112,908 backfile deposits, and 5,066,647 updated records. December recorded 1,142,863 added deposits, including 1,000,501 current and 142,362 backfile deposits, and 4,067,015 updated records.
Journal coverage and open access
Clarivate says the 2024 JCR covers more than 21,800 journals. About 5,800 of those journals publish all content via open access. Clarivate also says the 2024 JCR unified rankings cover 229 science and social science subject categories.
These figures describe JCR coverage and its ranking framework. They do not represent every journal in every field, nor do they establish that all journals outside JCR lack quality or influence. The open-access figure is specifically the number Clarivate identifies as publishing all content via open access among the journals covered by the 2024 JCR.
The contrast between Crossref’s 163,709 listed journals and Clarivate’s more than 21,800 JCR journals is a useful reminder that databases use different inclusion rules and purposes. Crossref is a DOI and metadata registration service; JCR is a citation-analysis and journal-evaluation product. Their counts answer different questions.
Metadata, volunteers, and report reach
DOAJ’s Annual Highlights 2024 was published on February 20, 2025. The report is 3.4 MB and had 1,290 total views on Zenodo, along with 662 total downloads. DOAJ says its achievement list includes 100+ volunteer editors and ambassadors.
Those DOAJ figures measure the publication and reach of one annual report, not the size of the DOAJ journal directory or the total number of open-access journals. They are useful for understanding how an open-access infrastructure organization reports its activity, but they should remain separate from journal-count statistics.
Taken together, the available measurements describe a layered system: Crossref records metadata and identifier relationships; PubMed counts citations; Clarivate reports coverage within JCR; and DOAJ reports activity around its open-access work. The most reliable comparisons keep each source label, geography or worldwide scope, and measurement period visible instead of collapsing all figures into one total.