A backlink is one link from a page on another website to your site. A referring domain is a unique website that links to you at least once. So if one blog links to you from 40 posts, you have 40 backlinks but only 1 referring domain. In the backlinks vs referring domains question, referring domains is usually the better number to watch, because it tells you how many different sites vouch for you.
Backlinks still matter as a diagnostic. The raw count tells you how a link profile is shaped, and a strange gap between the two numbers is often the first sign of a sitewide link, a scraper or a spam problem. My rule is simple. Track referring domains for growth, and read backlinks for clues.
Why Does Picking the Wrong Number Cost You?
I’ve seen site owners celebrate a jump from 2,000 to 9,000 backlinks. Then I open the report, and one directory’s footer is responsible for almost all of it. Nothing about their authority changed. They just got counted more times by one site.
The cost runs the other way too. People buy “1,000 backlinks” packages because the number sounds big, then wonder why rankings don’t move. If you judge link work by the backlink count, you reward the cheapest, most repetitive links. If you judge it by referring domains, you at least reward variety.
How Do Ahrefs, Semrush, Moz and Google Define Them?
These are tool metrics, not Google metrics. Each crawler counts a little differently, so the same site shows different totals in every tool. Here is how the vendors themselves describe the two numbers.
Ahrefs. Its backlink metrics documentation says the backlinks metric “is a total count of all backlinks pointing at a target”, and referring domains “is the count of unique websites that point at a target”. Ahrefs has also explained that it records only one version of a link per page, so a menu link and a body link to the same URL count once.
Semrush. Semrush’s guide on referring domains puts it plainly: if Website A links to your site twice, “then you have two backlinks from one referring domain.” Its Backlink Analytics reports list both totals side by side, with a separate Referring Domains report.
Moz. Link Explorer uses different labels for the same idea. It reports links and “linking root domains”, meaning the unique root domains that link to a site, page or subdomain. Same concept, different name.
Google Search Console. Google’s own Links report doesn’t use either phrase. It shows “top linking sites” and “top linked pages”, and Google’s Links report help page notes that data is grouped by root domain, duplicate links are combined, and the report shows only a sample of links.
The Common Mistake: Treating a Tool Metric as a Ranking Factor
Most comparisons stop at “referring domains matter more”. That’s fair, but it skips a step. Google never said it counts referring domains.
What Google does say is narrower. Its link best practices page opens with “Google uses links as a signal when determining the relevancy of pages and to find new pages to crawl.” Its ranking systems guide says Google has “various systems that understand how pages link to each other” and that PageRank is still one of its core systems.
The original idea behind PageRank is public. The PageRank patent (US6285999), filed by Lawrence Page in 1998, scores documents by the quality and quantity of citations from other documents. A patent shows what someone designed, not what runs in Google today. But notice the word quality. Neither “backlinks” nor “referring domains” measures quality on its own.
That’s the real frame. Referring domains are a rough stand-in for how many independent sources link to you. Backlinks are a rough stand-in for how many times you’re linked. Neither is the thing Google scores.
Four Conditions That Decide Which Number to Watch
- Where the links sit on the page. Links in footers, sidebars and blogrolls repeat on every page, which inflates backlinks without adding sources.
- How big the linking sites are. A 10,000-page site linking sitewide can add thousands of backlinks. A 20-page local blog can’t.
- What question you’re asking. Growth and competitive gaps are domain questions. Spam, scrapers and broken campaigns are backlink questions.
- Which tool you use. Ahrefs, Semrush and Moz run separate crawlers, so compare numbers inside one tool, never across tools.
Why Do Sitewide and Footer Links Inflate Backlink Counts?
A sitewide link is one that appears in a template, so every page of the site carries it. A web design credit in the footer, a “partners” block in the sidebar, a blogroll. If that site has 3,000 indexed pages, one decision by one webmaster can show up as 3,000 backlinks in your tool.
Google treats one version of this as spam. Its spam policies list “widely distributed links in the footers or templates of various sites” as a link spam example. That line is about the same link placed across many different sites, like a theme or widget credit, not a friend’s blog that happens to link to you from its sidebar.
In my experience, the honest single-site sitewide link isn’t a problem. It just isn’t worth thousands of links. I count it as what it is: one site’s endorsement.
The other common inflators are scraper sites that copy a page containing your link across hundreds of URLs, pagination and tag archives that repeat a post excerpt, and forum signatures. All of them push backlinks up. None of them add a new, independent source.
Referring Domains: Best for Measuring Growth
Best for: anyone tracking whether link building or digital PR is working month over month.
Sweet spot: comparing your site with the three to five pages that outrank you on a target keyword.
Strengths:
- It’s harder to inflate. One site stays one domain however many times it links.
- It maps better to the idea of independent sources vouching for you.
- It makes competitor gaps readable. “They have 180 referring domains, we have 40” is a useful sentence.
Weaknesses:
- It treats a national newspaper and a spammy expired domain as the same “1”.
- Tools disagree on what a domain is. Ahrefs says it counts separate user blogs on platforms like Blogspot as separate domains, so 50 spam subdomains can look like 50 referring domains.
Backlinks: Best for Diagnosing Problems
Best for: audits, spam checks and understanding why a profile looks odd.
Sweet spot: spotting the handful of domains responsible for most of your links.
Strengths:
- A sudden spike shows exactly when something changed, which is often the first clue in a negative SEO scare or a scraper outbreak, and the point where I’d run a step-by-step backlink audit before touching anything.
- It shows which of your pages earn links, not just which sites link.
- The backlinks-per-domain ratio exposes sitewide and template links fast.
Weaknesses:
- It rewards repetition, so it’s easy to game and easy to misread.
- It says nothing about whether the links come from different, independent sites.
Head-to-Head on the Criteria That Decide It
| Criterion | Backlinks | Referring Domains | Winner and Why |
|---|---|---|---|
| Resistance to inflation | Low, one sitewide link can add thousands | High, one site counts once | Referring domains |
| Tracking link-building progress | Noisy | Clear month-over-month signal | Referring domains |
| Spotting sitewide, scraper or spam links | Shows the spike | Hides it inside one domain | Backlinks |
| Seeing which of your pages earn links | Page-level detail | Site-level only | Backlinks |
| Comparing with competitors | Misleading | Useful inside one tool | Referring domains |
| Judging link quality | Doesn’t | Doesn’t | Neither, check the sites |
My Verdict on Backlinks vs Referring Domains
Referring domains is the better number for small businesses tracking link growth, because it counts independent sources instead of repetition. However, if you are auditing a profile or investigating a sudden spike, the backlink count is the right place to start.
When a client asks me which number to report, I say referring domains, filtered to sites that are real, relevant and indexed, which I test with the eight checks in my backlink quality scorecard. A site that went from 30 to 45 relevant referring domains in a quarter made genuine progress. A site that went from 800 to 4,000 backlinks tells me nothing until I see which domains caused it.
Honestly, both numbers are proxies. John Mueller once said in a 2021 hangout that the total number of links is “completely irrelevant” to Google, as quoted by Search Engine Journal. I read that as a warning against chasing either count. I cover what Google says about links in more depth in do backlinks still work.
When Does the Answer Flip?
There are cases where I’d look at backlinks first, or where neither count helps much.
- If a competitor’s backlink count is 50 times their referring domains, look at backlinks. A few sitewide or template links are propping up the total, and you don’t need to match it.
- If your backlinks jumped overnight but referring domains barely moved, go to the backlinks report and sort by domain. Usually it’s a scraper or a new footer link, not an attack.
- If you run a local business, neither number tells the full story. Citations and local press matter, and my local citations guide explains which ones count.
- If a tool shows lots of referring domains from subdomains of one platform, switch on the tool’s subdomain filter before trusting the total.
What Should You Actually Do With These Numbers?
Here’s the workflow I use in audits. It takes about 20 minutes in any major tool.
- Pull referring domains for your site and the top three competitors on one keyword, in the same tool.
- Divide backlinks by referring domains for each. A ratio far above your competitors’ usually means sitewide or scraper links.
- Sort referring domains by the number of links they send and open the top ten. Check where the link sits and whether the site is real.
- Compare relevant referring domains, not totals. Relevance beats volume every time.
If the gap is at the base level (no business profiles, no directory listings, no industry associations), that’s where I’d start. Our foundation link building covers brand profiles, business listings and niche directories only, which is the unglamorous layer most new sites skip. For how fast to add links after that, see my post on drip-feed link building.
One more point on paid links, since it comes up in every link conversation. Google’s policy is plain: paid links are allowed only if they’re qualified with rel="sponsored" (or rel="nofollow"). A paid link without that attribute is link spam under Google’s rules, however many referring domains it adds. If you’re not sure where your profile stands, start with a free SEO audit. The flip side of this topic, mentions without links, is covered in backlinks vs brand mentions.
Frequently Asked Questions
Is It Better to Have More Backlinks or More Referring Domains?
More referring domains, in most cases. Ten different relevant sites linking once each usually beat one site linking a thousand times, because the second case is really one source. Just remember that neither count measures quality, so check the sites themselves.
How Many Backlinks Should One Referring Domain Send?
There’s no correct ratio from Google. In practice, most editorial links arrive one to a few per domain. When a single domain sends hundreds, it’s almost always a template, sidebar or scraper link, and it should be read as one endorsement.
Why Do Ahrefs, Semrush and Search Console Show Different Numbers?
Each one crawls the web separately and counts differently. Ahrefs counts user blogs on platforms like Blogspot as separate domains, while Search Console groups data by root domain and shows only a sample of links. Compare trends within one tool, never totals across tools.
Are Sitewide Backlinks Bad for SEO?
Not automatically. A genuine partner or friend linking from a sidebar is fine. Google’s spam policies target “widely distributed links in the footers or templates of various sites”, which means the same credit link pushed onto many different sites, like a theme or widget footer.
Last updated: September 2026 by Mizanur Rahman
