stripe.comgithub.iocss-tricks.comlobste.rsalgolia.comweb.dev

referring domain vs backlink - the difference

a backlink is one link. a referring domain is one site - it can give you ten thousand links and still count as one. the two counts describe different aspects of a link profile; neither alone establishes link quality or ranking impact.

pete the seo wizard
· updated · 7 min read · 1,280 words
sharexlinkedin

a backlink is one hyperlink. a referring domain is a unique website that links to you. one referring domain can host one backlink or ten thousand backlinks - it still counts as one referring domain. this sounds like a pedantic distinction. it isn't. counting distinct sources helps separate broad coverage from one site repeating a link. neither count reveals google's weighting or establishes link quality without inspecting the context.

a backlink is a single <a href> from one page to another. if the new york times runs an article that links to your homepage, that's one backlink. if a forum thread mentions you in three posts and each one links to a different page on your site, that's three backlinks.

the page-level link is the atomic unit. every other backlink metric - referring domains, anchor text distribution, follow vs nofollow ratio - is a roll-up over backlinks.

what's a referring domain

a referring domain is the unique website that contains backlinks pointing at you. it's the registrable hostname - nytimes.com, notwww.nytimes.com nor cooking.nytimes.com nor a specific article url.

if every page on nytimes.com linked to your homepage, you'd have millions of backlinks but exactly one referring domain. that's the whole point of the metric: it normalizes for sitewide footprints and counts editorial votes from distinct websites.

quick definition for the rush

backlink = one link from a page. referring domain = one website (regardless of how many of its pages link to you). distinct sources describe breadth; repeated links describe volume. neither number is a published google ranking formula.

why distinct sources and repeated links tell different stories

if one site links to you 100 times, the total might reflect a repeated footer rather than 100 separate editorial decisions. links from 100 sites describe a broader set of sources, but those sources still need inspection. there is no universal exchange rate between repeated links and distinct domains.

a real example. crawlgraph's data shows vercel.com has roughly 5,000 backlinks but only 200 referring domains - a 25:1 ratio, which is normal for a developer-tools company with a lot of repository readme references and forum cross-links. compare to a site with 5,000 backlinks from 20 referring domains (250:1 ratio): that's a sitewide-link footprint - likely a directory or two repeating the same link on every page. google reads those very differently.

breadth is only one part of the review. a relevant editorial link and an unrelated template link can each add one referring domain without offering the same value. the backlink quality checklist explains how to inspect the source page, context, destination and link attributes before treating either count as evidence of a useful link.

the backlinks-to-referring-domains ratio

ratio matters more than either number on its own. as a rough guide:

ratio (backlinks : referring domains)signal
under 5:1healthy. each domain links to you a few times - typical editorial pattern.
5:1 to 15:1normal for sites with sticky content. blog posts get cited on multiple pages of the same site.
15:1 to 50:1elevated. usually means a few high-volume domains dominate (template footers, sitewide navigation links).
50:1+footprint signal. google's spam systems will look at this. legitimate sometimes (e.g., software with many doc pages linking to a homepage), but worth investigating.
100:1+almost always a paid network or a sitewide-spam pattern.

edge cases: subdomains, www, and what counts as one domain

every backlink tool collapses these slightly differently:

  • example.com and www.example.com are always treated as one referring domain.
  • blog.example.com vs example.com: ahrefs and most tools treat them as one referring domain (registrable domain). moz can split them in some reports. crawlgraph rolls them up to the registrable domain.
  • example.co.uk and example.com are different domains - they're different registrations.

how to find your own referring-domain count for free

three options, no subscription required:

  1. google search console - links report shows top linking sites, but capped and not always complete. own-site only.
  2. bing webmaster tools - backlinks report. own-site only, generally less data than ahrefs but free.
  3. crawlgraph free tier - first 5 referring domains for any site, including competitors. for the full list, $99 once. see five free ways to find backlinks for the full breakdown.

what the 2024 google leak revealed

the 2024 leak of internal google search documentation confirmed several link-related attributes that ranking-conscious seos had long suspected. relevant here: numUniqueLinkingDomains appears as a real attribute on documents - google explicitly counts unique domains, not just total links. a separate attribute, linkDecayBoost, modulates the contribution of a link based on how recently it was crawled.

the takeaway: the difference between backlinks and referring domains isn't a third-party-tool quirk. it's baked into google's ranking systems. focusing on referring domain growth - not raw backlink counts - is the empirically correct read.

faq

are referring domains more important than backlinks for seo?

referring-domain count tracks breadth, while backlink count measures volume. neither alone establishes ranking value, and google does not publish a simple weighting between them. inspect the linking pages and relevance before judging a change in either metric.

can one site count as multiple referring domains?

usually no. the registrable domain - example.com - is the unit. some tools split certain subdomain patterns (mostly moz on user-generated platforms), but the modern norm is one site = one referring domain.

how many referring domains do i need to rank?

depends on the keyword and competing pages. for most non-commercial long-tail queries, 5-15 quality referring domains can crack page one. for competitive commercial terms ("free backlink checker," "best crm"), top-10 pages typically have 200-2,000+. checking the top 10 results for your target keyword in any backlink tool is more useful than chasing a number in the abstract.

what counts as a "high-quality" referring domain?

a referring domain on a topically-relevant site, with its own healthy link profile, that links to you in body content (not in a sidebar or footer). a single link from a topically-relevant dr-50 site usually outweighs a dozen links from unrelated dr-80 sites.

ahrefs · backlinkslocked
upgrade required · $129/mo
crawlgraph · live $99 once
G
github.io92
C
css-tricks.com88
L
lobste.rs86
A
algolia.com84
W
web.dev80
same data · one-time
$99oncevs $129/mo
unlock the data →
stripe checkout · instant access
guides#definitions#ranking factors#link profile
sharexlinkedin
pete the seo wizard
author

writes the queries we run internally. ships one tactical post a week.

the dispatch
one email a month.

plus one when a new common crawl release lands. that is all.