What DR actually predicts

What DR is calculated from

Domain rating is an Ahrefs metric. It scores a domain from 0 to 100 on a logarithmic scale, derived from the links that point at it: how many unique domains link to the site, how strong those linking domains are in the same model, and how many sites each of them links out to in turn. A link from a domain that links to almost nobody carries more weight than a link from one that links to everybody.

Two consequences follow from the logarithmic scale, and they are the ones people forget. Moving a site from the low end to the middle takes far fewer referring domains than moving it from the middle to the high end. And a gap of ten points near the top represents a much larger difference in link volume than the same ten points near the bottom.

Notice what is not in that calculation. Traffic is not in it. Topical relevance is not in it. Rankings are not in it. Whether the site publishes anything a human would read is not in it. DR is a model of one thing — referring-domain strength — and it is honest about being that one thing. The confusion is ours, not the metric's.

What it correlates with, and how strongly

DR tracks the size and strength of a link profile, because that is what it measures. It moves with traffic and rankings as well, in the ordinary way that established sites tend to accumulate both links and audiences over time. A site that has been publishing usefully for years will usually have earned links and usually have readers. The two travel together often enough that DR starts to look predictive.

But that is a shared cause, not a mechanism. Nothing about a high DR makes a site rank, and nothing about ranking requires a high DR. It is not a Google metric. It is a third-party model of the link graph, published by Ahrefs, and search engines do not take it as an input.

We are not going to put a figure on the strength of the relationship, and you should be wary of anyone who does. Any such number depends on which set of sites was measured, over what period, in which market — and none of them generalise to the domain in front of you. What we will say is this: because the association is a by-product rather than a cause, DR on its own is not enough to decide a placement.

Where it misleads: the traffic-free high-DR site

The clearest failure case is the site with a strong DR and effectively no organic traffic. Once you start checking for it, you find it, and there are only a few ways one gets made.

The first is history. A domain earned real links years ago, then stopped publishing, changed hands, or lost its rankings to a competitor with fresher content. The links persist because links persist. DR reflects a link profile the site earned in a different era and describes nothing about it today.

The second is a rebuilt expired domain. Someone registers a domain that once had authority, puts a thin site on it, and sells placements against a link profile they did not earn, pointing at content that no longer exists.

The third is deliberate acquisition — private networks, reciprocal arrangements, bulk purchases. Any metric derived from the link graph can be moved by someone willing to build links into it, and DR is no exception.

In all three cases the number is real and the value is not. Nobody visits the page your link sits on. The link comes from a site with no ranking presence in your topic, and possibly no ranking presence at all. You have bought a number.

To be fair to DR: it is not lying in any of those cases. It is reporting referring-domain strength accurately. The mistake is ours, for asking it a question it was never built to answer.

The metrics we would check first instead

Start with organic traffic, and specifically the trend rather than the snapshot. A flat line near zero on a high-DR site tells you almost everything you need. A declining line tells you the site is losing whatever it had. A stable or rising line tells you search engines currently send it people.

Then look at the keyword profile and ask whether it overlaps with yours. A site ranking for terms in your niche is a site whose audience and topical signals sit next to your page. A site ranking for nothing but its own brand name, or for a scatter of unrelated informational terms, is not.

Then check whether it ranks for anything commercially meaningful — terms someone would search with an intent to act. That is a harder bar than traffic volume, and it is the one that separates sites with real search presence from sites with a large archive of low-competition posts.

Then look at where the referring domains come from, not only how many there are. Links from sites covering the same subject behave differently from the same number of links out of a network.

Finally, check whether the topic already exists on the site. A placement lands better on a domain that was covering the subject before you arrived, because that is the difference between a link that belongs on the page and a link that was inserted into it.

This is why our listings carry domain rating alongside organic traffic, referring domain count and spam score, refreshed periodically, rather than DR on its own. We are not asking you to ignore DR. We are asking you to stop treating it as a summary of everything else, because it never was one — and if a site's only strong number is DR, that is information too.