Every week someone asks me what their Domain Authority is, as if the answer will tell them something useful. It won’t. DA is a metric invented by a software company, not a signal that Google uses, and the amount of collective anxiety directed at it is wildly disproportionate to its actual importance.
The SEO industry is good at this — taking something either trivial or irrelevant, building a number around it, then selling services to track and improve that number. GEO is the latest example. Domain Authority has been at it since 2008.
Here’s the short version: Moz made it up. You can stop now.
What Domain Authority actually is
Moz introduced Domain Authority in 2008 as a proprietary score from 0 to 100, designed to predict how well a domain might rank in search results. It’s calculated from Moz’s own index of the web — the backlinks they’ve crawled, the quality scores they’ve assigned. It’s a number that describes how Moz sees your site, not how Google sees it.
Google does not use DA. Google does not acknowledge DA. Google has its own internal version of link authority (PageRank, which is updated continuously and never published) and it works on signals that no third-party tool has direct access to. DA correlates with rankings only in the way that height correlates with shoe size — the two often move together, but one doesn’t cause the other, and there are plenty of exceptions.
Ahrefs has their own version, Domain Rating (DR). Semrush has Authority Score. They’re all different numbers, all built on different crawls of the web, all measuring slightly different things. None of them are what Google uses.
How the obsession plays out
DA has become the currency of cheap link building, and that’s where the real damage happens.
If you’ve ever looked at a guest post marketplace or a link insertion service, you’ll have seen sites sold by their DA score. “DA40+ links from $60.” Agencies set DA thresholds for their outreach: only target sites above DA30, only build links from DA50+. Clients get monthly reports where DA moving from 24 to 27 is presented as a win.
What this produces is a market for DA manipulation. If links from high-DA sites are valuable, and DA is based on the number and quality of backlinks you have, then buying cheap links from other high-DA sites temporarily inflates your score. You can buy a DA50 that’s built from junk. The metric that was supposed to signal quality becomes a thing you game, which makes it less useful as a signal, which nobody involved in the transaction has any incentive to mention.
A high DA site can rank for nothing you care about
The real test of whether a site is worth getting a link from is whether it has a real audience reading relevant content. A DA60 site built for link selling, with no editorial standards and no real traffic, is worth less than a DA20 blog with 3,000 readers who actually care about your industry.
Check this yourself. Take any high-DA site you’ve seen offered in a link marketplace. Open it in Ahrefs or Semrush and look at the organic traffic graph. You’ll often find a site with a high score and near-zero organic traffic — a site Google has already deprioritised, still carrying the numerical remnant of links it received years ago.
DA is a lagging indicator of past behaviour. Google is running on current signals. Those are not the same thing.
What Google actually cares about
Google’s link quality assessment comes down to a few things: is the linking site relevant to your content, does it have real editorial standards, does it attract real readers, and is the link placed in a context that makes sense? A mention in a real article, on a real site, read by people who care about your topic, is valuable. A link in a footer of a directory nobody visits is not, regardless of what Moz thinks the domain is worth.
Relevance matters more than authority score. A DA30 trade publication that covers your exact niche is a better link than a DA70 generic news site where you got buried in a roundup nobody reads.
What you should actually track
If you want to understand whether your backlink profile is improving, look at:
- Actual rankings for target keywords. Do you rank higher than you did six months ago? That’s the question.
- Organic traffic. Is more of it arriving, from the right queries?
- The quality of individual links. Use your own judgement: is this a real site with a real audience? Would you be happy for a potential customer to see you mentioned there?
If you’re building links and want a rough proxy for quality, look at the linking site’s organic traffic in whatever tool you use — not its DA score. A site with 10,000 monthly organic visitors is real. A site with 200 visitors and a DA of 55 is a yellow flag.
What to actually do
Stop using DA as a quality benchmark when building or evaluating links. Look at the actual site. Does it publish content people read? Is it indexed? Does it have organic traffic? Is it relevant to what you do?
If you’re getting a monthly SEO report and DA is the headline metric, ask what your actual rankings look like. DA going up is not the same as rankings going up. Traffic going up is not the same as revenue going up. Know what you’re measuring and why.
The number exists because tools need to sell subscriptions and clients need something legible to look at. That’s fine. Just don’t confuse a dashboard metric with a strategy.
