Rankfield

Coverage and dates

More ▾

Understand what the Rankfield web graph covers, how old the data is, and why a domain or a link can be missing.

What the graph holds

The graph is built from public web crawl data. It records a link from domain A to domain B when that data includes a page on A that links to B. The current release has about 118 million domains and 2.8 billion domain-to-domain links, from crawls of May, June and July 2026 (Get Dataset Status gives the exact figures).

How fresh it is

Every answer reflects the release, not a live check of the web. Links added or removed after the crawl show up in a later release. Answers name their release in crawlRelease and asOf where the format allows.

Why something is missing

  • A domain answers "inGraph": false, with zeros: the crawl data has no page that links to it, or it was not in this release. It does not mean the site is offline or has no links, and the zeros are not a measurement.
  • A count is null: the domain is in the graph, but its count is unavailable. Treat it as unknown, not zero.
  • A linking domain is not listed: the crawl data does not include the page that links. Crawl frequency, robots.txt rules and page discovery all limit what a crawl sees.

Domains, not pages

The graph links domains to domains. It has no source page URLs, anchor text or nofollow flags. To see the links on a live page, use Extract links from a page.