Referring Domains vs Backlinks: Why the Domain Count Is the One to Read (2026)
A referring domain is one site that links to you, however many times it does. Why the referring domain count beats the raw backlink count, how Ahrefs, Semrush, Majestic and Moz each define it, why every listing in the directory shows a referring domain band, and how to read that band against a domain authority score.
A referring domain is a website that links to the site you are looking at. A backlink is one link. One site can give you a thousand backlinks and still be one referring domain. That is the whole distinction, and it is the reason the referring domain count is the first number to read on any link profile and the raw backlink count is the last.
This page covers the definitions the main tools use, why the two counts diverge so much, why the Revised directory shows a referring domain band on every listing, and how to read that band against a domain authority score. How backlinks work covers what a link is and why search engines count them at all.
What is a referring domain?
Ahrefs' help center defines it with an example that is hard to improve on: "Referring domains are domains from which the target website or web page has one or more backlinks. For example, if a web page has a backlink from the New York Times, then it has one referring domain. If it has a link from the New York Times and Forbes, that it has two referring domains. If it has two backlinks from the New York Times, then it still has one referring domain."
Semrush's glossary entry says the same thing shorter: "A referring domain is an external website that links to a specific website or webpage." Moz uses the phrase linking root domains for the same count, and its Domain Authority page says "a referring domain is a unique domain that links to your website." Majestic's glossary lists Referring Domain as a term and its Site Explorer has a Ref Domains tab beside the Backlinks tab. Four vendors, one definition, four labels. The unit is the linking site, counted once.
Why referring domains beat the backlink count
Because a backlink count measures the linking site's page count, not its opinion of you. A forum that puts your link in a footer has as many backlinks to you as it has pages. A blog that mentions you once in an article has one. The forum's count is a fact about the forum's template. The blog's count is a fact about an editorial decision. Referring domains count decisions.
Semrush's referring domains report page makes the practical claim that a site with fifty backlinks from fifty domains could be seen as more trustworthy than a site with five hundred backlinks from five domains. Ahrefs says the Domain Rating formula is not disclosed but that it is driven by "the number of domains linking to a website (referring domains)," the DR of those domains, and "the number of websites the referring domain links to." Moz's DA page lists "the number of linking root domains" first among its inputs. Every one of the composite scores starts from the domain count, not the link count, for the reason above.
Google does not publish a formula, so the only primary text is the negative one. Its spam policies define link spam as "the practice of creating links to or from a site primarily for the purpose of manipulating search rankings." A thousand sitewide links from one domain is the easiest kind of link to create and the easiest kind to discount. Fifty separate sites choosing to link is not something you can generate from one template.
The referring domain count also fails in a predictable direction, which makes it easier to read. It can be inflated by a network of throwaway sites, and that shows up as many domains with no history and no links of their own. It is rarely deflated. A site with 200 referring domains had 200 sites link to it. How to choose backlinks covers the second question, which is whether those 200 sites are worth anything.
Why do the counts differ so much between tools?
Because every tool counts its own index, and no index is the web. Ahrefs, Semrush, Majestic and Moz each run a crawler and each hold a different sample of pages. A site can show 340 referring domains in one and 190 in another without either being wrong. Our backlink checker reads from our own index, most of which is the Common Crawl web graph, an open crawl published several times a year, plus our own crawls of pages that link to the domains we track. It is smaller than a commercial index and it is built for expired-domain work rather than exhaustive coverage. A thin result there means thin in our index, not no backlinks.
The rules for de-duplicating also differ. Whether two subdomains of the same host are one referring domain or two, whether a link that redirects counts against the source or the target, and whether nofollow links are counted at all vary by vendor. Ahrefs says nofollow links do not pass DR. The backlink checker shows the link type per row so you can see the dofollow share yourself. Dofollow vs nofollow links has the detail.
| Tool | Label for the site count | Label for the link count | What feeds its composite score |
|---|---|---|---|
| Ahrefs | Referring domains | Backlinks | Referring domains, their DR, their outlinks |
| Semrush | Referring domains | Backlinks | Authority Score, from link quality and quantity |
| Majestic | Referring domains, Ref Domains | External backlinks | Trust Flow and Citation Flow, from link graph |
| Moz | Linking root domains | Inbound links | Domain Authority, linking root domains first |
| Revised | Referring domains | Backlinks | Revised Authority, from the Common Crawl web graph |
Compare a domain's count across tools by ratio, not by absolute number. If it has twice the referring domains of a competitor in one tool, it will usually have roughly twice in another. If the raw number is what you need, say which index it came from.
Why every listing in the directory shows a referring domain band
Every listing in the Revised directory shows a referring domain band, in five steps: under 10, 10 to 50, 50 to 100, 100 to 250, and 250 or more. The count comes from the Common Crawl web graph, de-duplicated to the registrable domain, and it is the first metric on the row because it is the first thing you would check by hand.
It is a band rather than an exact count for two reasons. The exact count changes between crawls and would be stale by the time you read it, and an exact count is a fingerprint. Names in the directory are masked until you reveal them, and a listing that said 187 referring domains could be matched against a backlink index to find the name without a reveal. A band cannot. The same logic applies to the other metrics on the row, which is why age is a year range and the archive record is a snapshot count and a span of years rather than dates.
The band is the filter that removes most of a raw drop list. Nearly every name on a daily deletion report has no linking sites at all, and the roughly 11,000 listings in the directory are the ones that still had some. Setting a floor at 10 to 50 and reading the named sources beside it is the check that how to find expired domains with backlinks walks through, applied before the names were listed.
How to read referring domains against domain authority
A domain authority score, whichever vendor's, is a 0 to 100 estimate of link strength that already has the referring domain count inside it. Reading the two together tells you something the score alone hides. Domain Authority vs Domain Rating vs Revised Authority covers what each score measures and why Moz says "Domain Authority is not a Google ranking factor and has no effect on the SERPs." What follows is how to read a score against the count.
| Referring domains | Authority score | What it usually means |
|---|---|---|
| High | High | Many sites link to it and some of them are strong. The profile you want |
| High | Low | Many weak sites link to it. Directories, profiles, a network. Read the sources |
| Low | High | A few strong links. One university or newspaper citation. Fragile but real |
| Low | Low | Nothing to see. Most of any drop list |
The high count with a low score is the case to be careful with. Across the directory, a high referring domain count correlates with a lower authority score more often than you would expect, because the domains with hundreds of linking sites are often the ones that were listed in every directory and scraped by every aggregator. The score is low because those linking sites have no links of their own. The named sources on the listing settle it. A count of 250 or more with sources that are all directories is a different asset from 50 to 100 with an encyclopedia and a government body among them.
The low count with a high score is the reverse. One link from a strong site moves a link-graph score a long way. That link is real, but if the page that carries it is removed the score goes with it. Evaluate an expired domain's link history covers checking whether the linking page still exists and whether anyone maintains it.
Run the two checks yourself on any domain with the domain authority checker for the score and the backlink checker for the referring domains and the named sources. For the set that has already passed both, the high authority expired domains hub lists the directory cut at a Revised Score of 65, and every row still shows its band so the fragile cases stand out.
What is a good number of referring domains?
There is no threshold, because the count only means something next to the competition for a query. Semrush's referring domain guide reports from its ranking factors study that pages ranking at position 10 have "fewer than 80 referring domains on average," and pages at position 1 have more. That is an average across every query, and the number for a local trade query is a fraction of it while the number for a finance query is a multiple.
For an expired domain the useful comparison is not a query but the drop list. A dropped domain with 10 or more referring domains is already in a small minority. One with 50 or more, from sources that were not directories, is rare enough that it is usually caught within seconds of the drop. That is why the band and the named sources are on every listing and why the directory re-checks that each name is still unregistered.
FAQ
What is the difference between backlinks and referring domains? A backlink is one link from one page. A referring domain is one linking website, counted once no matter how many pages on it link to you. Ten links from one blog are ten backlinks and one referring domain.
Why are referring domains more important than backlinks? Because each referring domain is a separate site's decision to link, while backlinks from the same site are usually a template repeating one decision. Every major authority score is built from the domain count first.
Why does Ahrefs show a different number of referring domains than Semrush? Each tool counts its own crawl, and the crawls hold different pages. De-duplication rules for subdomains, redirects and nofollow links also differ. Compare by ratio between domains inside one tool rather than by absolute number across tools.
Is referring domains a Google ranking factor? Google does not publish its ranking inputs, and no tool's count is read by Google. Google's spam policies do say that links created to manipulate rankings are spam, and a count of separate sites choosing to link is harder to manufacture than a count of links.
How many referring domains should an expired domain have? Enough that the named sources include sites that chose to cite it. In the directory, the 10 to 50 band and above is where that starts, and the sources beside the band tell you whether the count was earned.
To read referring domain bands beside the named sources and the score on names that have already dropped, browse the directory of expired domains with backlinks. Sort by band, open a listing's sources, and reveal a name only when the count was earned by sites you would want linking to you.