Common Crawl publishes individual web crawls and separate composite web graphs built from several crawls. crawlgraph needs another ingest and verification step before a published graph can answer backlink queries here. the live release list below shows both states, so a newer upstream graph is never mistaken for an indexed crawlgraph release.
the cadence in one sentence
individual crawl archives use IDs such as CC-MAIN-2026-39; composite web graphs use IDs such as cc-main-2026-jul-aug-sep. the graph follows its source crawls, and crawlgraph indexes a graph only after its own build and verification. always check the release lists below for current availability rather than assuming a fixed publication lag.
recent web graph releases
this table reads Common Crawl's published graph list. the complete archive is at graphinfo.json. domain counts and links describe the upstream graph, not what crawlgraph has indexed.
crawlgraph's newest queryable domain graph is cc-main-2026-apr-may-jun. Common Crawl's newest published web graph is cc-main-2026-jul-aug-sep; it is not yet queryable on crawlgraph. Check Common Crawl's web graph release list for the upstream source. Published and queryable are different states.
| published graph | source crawls | domain nodes | domain links | crawlgraph |
|---|---|---|---|---|
cc-main-2026-jul-aug-sep | CC-MAIN-2026-30, CC-MAIN-2026-34, CC-MAIN-2026-39 | 133,241,980 | 2,147,955,520 | not queryable here |
cc-main-2026-jun-jul-aug | CC-MAIN-2026-25, CC-MAIN-2026-30, CC-MAIN-2026-34 | 119,722,885 | 2,450,405,793 | not queryable here |
cc-main-2026-may-jun-jul | CC-MAIN-2026-21, CC-MAIN-2026-25, CC-MAIN-2026-30 | 117,963,409 | 2,774,625,913 | not queryable here |
cc-main-2026-apr-may-jun | CC-MAIN-2026-17, CC-MAIN-2026-21, CC-MAIN-2026-25 | 121,091,933 | 3,902,808,757 | queryable |
cc-main-2026-mar-apr-may | CC-MAIN-2026-12, CC-MAIN-2026-17, CC-MAIN-2026-21 | 118,760,321 | 4,343,610,896 | not queryable here |
the calendar uses commoncrawl.github.io/cc-crawl-statistics for source-crawl history. crawlgraph's Apr-Jun 2026 composite, its source crawl ids, and domain-graph totals are in the Apr-Jun 2026 release notes.
the warc-to-graph lag
the raw web crawl (warc files) is one product. the hyperlink graph derived from it is a separate product. graph construction follows the source crawls: links are extracted, hosts are normalized, and the published graph contains node and edge data at both host and domain level.
publication dates vary. the upstream graph list shows what exists now; crawlgraph then needs to build and verify its own query database. a link missing from one graph may have been missed by that crawl and does not prove the live web page has no link.
the news crawl runs faster
common crawl also publishes a news crawl with a different cadence. warc files there publish within hours of being written - useful for time-sensitive research on news sites, but irrelevant for general-web backlink intelligence. the news crawl is decoupled because the monthly schedule was, in their words, "not well-adapted to news content."
when's the next release
common crawl does not promise a fixed hyperlink-graph date. crawlgraph cuts over only after the next composite has been ingested and verified; the homepage's index-updated line and the release notes identify the active snapshot after that cutover.
why the cadence matters for seo data freshness
every backlink tool depends on what its source observed. ahrefs and semrush maintain their own indexes; crawlgraph uses published Common Crawl graphs. a newly published site can be absent from a snapshot taken before it was available to crawl.
if you need to confirm a link placed today, inspect the source page directly. use crawlgraph to research referring domains observed in the named graph release, and compare named releases to see which observations changed.
when you query crawlgraph for a domain, the data comes from the active, verified DuckDB release artifact. repeat lookups can be served from the SQLite result cache. see how the underlying graph works and five free ways to find backlinks.
faq
how often does common crawl release new data?
crawl archives and web graphs have separate publication schedules. individual crawl IDs look like CC-MAIN-2026-39; composite graph IDs look like cc-main-2026-jul-aug-sep. use the source lists above for the current releases.
when will the next common crawl release be?
common crawl does not publish a guaranteed hyperlink-graph date. crawlgraph announces a new active composite only after its artifact has been ingested and verified.
what's the difference between common crawl's main crawl and news crawl?
the main crawl runs every 1-2 months and covers the general web. the news crawl publishes warc files within hours of being written and only covers news sites. the news crawl was decoupled because the monthly main-crawl cadence is too slow for news content.
how long after a crawl is the hyperlink graph available?
there is no guaranteed interval. a graph is built from several crawls, then crawlgraph separately indexes and verifies it. compare the source crawl IDs in the table with crawlgraph's queryable state before making a freshness claim.
writes the queries we run internally. ships one tactical post a week.
plus one when a new common crawl release lands. that is all.