Backlinks vs Referring Domains: Why the Scores Disagree

You put your domain into four tools and got four different numbers. One says you have 8,400 backlinks, another 3,100. Your authority score is 31, or 44, or 19, depending on which tab you are looking at.

The natural conclusion is that three of them are broken. The actual situation is worse and more useful: backlinks and referring domains are two different counts of two different things, the authority scores on top of them are opinions rather than measurements, and none of those scores is a number Google has ever published.

So here is what each figure means, why they disagree by design, and which of them is worth putting in a report.

Backlinks and referring domains are not the same claim

A backlink is one link on one page. A referring domain is one website that links to you at least once, however many times it does it.

The gap between the two is where most of the confusion lives. Say a directory adds you to its footer and that footer appears on 9,800 pages. Your backlink count jumps by 9,800. Your referring domain count goes up by one.

Ten thousand links from twelve domains is twelve votes, not ten thousand. And the direction of that ratio is the single fastest read on a link profile you will get:

• Thousands of links, dozens of domains. Sitewide footers, blogroll widgets, a template someone shipped with your link in it. One relationship counted many times.
• Hundreds of links, hundreds of domains. Roughly one link each — the shape of a profile built from people mentioning you individually.

Neither shape is automatically good or bad. But a headline backlink number with no domain count beside it is close to meaningless, and it is the number every vendor puts in the biggest font.

Why the raw counts disagree between tools

Even the countable figures differ, and that part is not anybody being dishonest.

No tool has Google’s index. Each one maintains a link index of its own — a crawl of some portion of the web, refreshed on some schedule, with its own rules for how long a link stays in the set after it stops being seen. Two indexes built that way will always disagree at the edges: different pages discovered, different recrawl intervals, different decisions about when a dead link finally drops out.

Which is why comparing a backlink count from one tool this month against another tool’s from last month tells you nothing about your link building. Compare a tool against itself over time, the same way you would with rank tracking, and for the same reason: consistency of instrument is what makes the trend readable.

Authority scores are vendor opinions

Domain Authority, Domain Rating and every “authority score” in the category are proprietary metrics invented by the companies selling them. They are model outputs, each built on its own link index and its own formula, and they are not calibrated against each other. Disagreement isn’t a bug in one of them; it is the necessary result of four companies estimating the same unpublished quantity in four ways.

More importantly, none of them appears in Google’s documentation, because Google does not publish a site-level authority number. What its link best practices guide does say is narrower and more useful: “Google uses links as a signal when determining the relevancy of pages and to find new pages to crawl.” A signal, for relevancy, among the hundreds of factors its ranking systems weigh.

That is the whole documented position. Everything beyond it — a two-digit score, a logarithmic curve, a colour-coded badge — is a vendor’s attempt to compress something Google keeps to itself into a number you can put in a slide.

Ours is an estimate too

This is the part where a post like this usually gets quietly selective, so: Reachavo shows an authority score in Backlink Analysis, and it belongs in exactly the same bucket. It is our model, over our link data, with our assumptions. It is useful for ranking a list of prospects against each other and it is not a fact about a domain.

Any vendor telling you their score is the accurate one is selling you the thing this whole post is about. The honest version of the claim is that a consistent estimate is a decent way to sort candidates, and a poor way to set a threshold.

What to read instead

Four things, all countable, none of them a composite score.

Those last two are worth dwelling on, because they are how a profile stops looking earned. Google’s spam policies define link spam as “the practice of creating links to or from a site primarily for the purpose of manipulating search rankings”, and the examples are specific: buying or selling links for ranking purposes, “excessive link exchanges”, and “advertorials or native advertising where payment is received for articles that include links”.

Worth knowing that paying for placement is not itself the problem. Google treats buying and selling links as “a normal part of the economy” when the links are qualified with rel="nofollow" or rel="sponsored". The policy is about undeclared ranking manipulation, not about commerce.

So what is a good number?

There isn’t one, and that is the honest answer rather than an evasive one. A score of 40 means something different in a niche where the leaders sit at 45 than in one where they sit at 85.

The usable version of the question is comparative. Pull the same metric, from the same tool, for yourself and the three sites currently outranking you for the terms you care about. The gap between you and them, in referring domains and in the relevance of those domains, is a real target. An absolute threshold copied from a blog post is not.

This is the same discipline that makes a keyword gap useful rather than overwhelming — measure yourself against the specific competitors on the specific queries, which is the argument in the post on keyword gap analysis.

The short version

Referring domains counts relationships; backlinks counts mentions of those relationships, and one sitewide footer can carry ten thousand of them. Tool counts differ because every link index is a different partial crawl of the web. Authority scores differ because they are competing estimates of a number Google has never published — ours included. Track referring domains over time, look at who they are and how they describe you, and compare against the sites actually outranking you rather than a threshold.

If you want that view without stitching four exports together, Backlink Analysis reports referring domains, anchor distribution and follow ratio side by side — with the score clearly labelled as the estimate it is.