Where our data comes from

You should be able to check our work. Here is exactly what powers every number on this site.

The source

All link data comes from Common Crawl, a non-profit that has crawled the open web since 2008 and publishes the results freely. We use their domain-level hyperlink graph, release cc-main-2026-may-jun-jul, plus the previous release for change detection.

• 117,963,409 domains
• 2.77 billion domain-to-domain links in the current release
• 7.11 billion across the two releases we hold

How authority is computed

Common Crawl publishes harmonic centrality for every domain — a measure of how close a node sits to every other node in the graph. It is an established metric in network science, not something we invented.

We convert its rank to a 0–100 score on a logarithmic curve: score = 100 × (1 − ln(rank) / ln(N)), where N is the number of ranked domains. The log curve matters because the gap between rank 10 and rank 100 is enormous, while the gap between 1,000,000 and 1,000,090 is noise.

What we deliberately don't claim

• This is not Moz DA or Ahrefs DR. Those are proprietary scores on proprietary crawls. Ours will correlate, but it will not match.
• It is domain-level. One referring domain counts once, however many pages it links from.
• It is not real time. Data refreshes when Common Crawl publishes — every few months, not continuously.
• Coverage is broad, not complete. Common Crawl samples the web widely; it does not fetch every page of every site.

How to compare our numbers with another tool's

If our score for a domain differs from Ahrefs' DR or Moz's DA, neither number is “wrong” — they are measurements of different indexes. Every authority metric is a function of the crawl behind it, and no two crawls see the same web. A domain with a strong score in one tool and a weak one in another usually means its links sit in a part of the web that one crawler sampled more heavily than the other.

So use any authority score the way it was designed to be used: as a relative comparison inside one tool. “This site scores 20 points higher than that one on the same index” is a meaningful statement. “This site is a 46” on its own is not, and mixing scores from two tools in the same sentence is meaningless. If you are tracking progress over time, track it on one metric and stay there.

What the score does not tell you

• It is not a Google ranking factor. Google does not use it, and no third-party score is an input to search results. It is an estimate of link strength, nothing more.
• It ignores relevance. A strong link from an unrelated site can be worth less than a modest one from directly within your topic. The score cannot see that.
• It says nothing about content or traffic. Plenty of high-authority domains get almost no search traffic, and plenty of small sites out-rank them on the queries that matter.
• It can be gamed at the low end. Cheap links move a small score noticeably and a large one barely at all — which is precisely why a rising score is not proof of progress.

Why we spell this out

Most SEO metrics are black boxes, which lets vendors imply more precision than they have. Our source is public and our formula is one line, so you can reproduce any number here yourself. We would rather be checkable than impressive.