Google's own John Mueller has stated it directly: "Google does not use Domain Authority or any third-party metric in our algorithms." And yet Domain Authority correlates with rankings, consistently, across every study that's measured it. Both statements are true at the same time, and understanding why is more useful than treating DA as either a scam or a scoreboard.
What Domain Authority Actually Is
Domain Authority is a prediction score built by Moz, an SEO software company, on a 1-100 logarithmic scale, designed as a comparison tool rather than a replica of Google's algorithm. Ahrefs' equivalent is called Domain Rating (DR); Semrush's is Authority Score. All three are calculated primarily from backlink data — the number of linking root domains, total backlinks, and the estimated strength of the sites doing the linking — using each company's own proprietary model and web index. Google has no access to any of these numbers and no reason to use them, since they're built entirely outside Google's own systems.
Why It Correlates With Rankings Anyway
The mechanism is straightforward once stated plainly: DA and DR are built almost entirely from backlink data, and backlinks are a real signal Google does reward. High-DA sites tend to rank well not because Google checks their DA score, but because both DA and rankings are downstream of the same underlying input — a strong backlink profile. One clear analogy: DA works like a thermometer. The thermometer doesn't create the heat; it reports it. Your link profile creates the ranking power, and DA reports its size after the fact.
How Strong Is the Correlation, Actually?
This is where the honest answer gets more nuanced than either "DA is meaningless" or "DA predicts rankings." Onely's correlation study across 2,000 SERPs found both DA and DR show a modest but real correlation with rankings — strongest on medium-difficulty keywords, weaker at the extremes. A separate analysis reported a correlation coefficient in the range generally considered statistically weak-to-moderate, not the strong predictive relationship the marketing around these metrics sometimes implies. Ahrefs' own correlation study of over 218,000 domains found domain-level authority correlates more meaningfully with rankings than page-level metrics (like Page Authority or URL Rating) do — a useful distinction, since a site's overall authority and one specific page's ranking potential aren't the same question.
DA Is Trivially Gameable, Which Undercuts Its Reliability Further
A concrete illustration matters more here than an abstract warning: services openly advertise DA and DR boosting starting around $15, with results delivered within days. Documented test domains have reached DR 44–59 and DA 50+ through purchased links alone — while real organic traffic on those same domains stayed at zero. Google's own ranking systems have caught and penalized sites like this, with traffic dropping to zero even as the third-party authority score kept climbing, because Google's internal signals detect what a purchased-link "zombie domain" looks like in ways a public, link-count-based metric structurally cannot.
Different Tools Don't Even Agree With Each Other
Beyond the Google-doesn't-use-it point, the tools disagree with one another in ways that matter if you're using any single score as a hard benchmark. Semrush's Authority Score commonly runs 15–20% lower than an equivalent Moz DA score for the same domain. A domain-level academic reliability study comparing Moz, Semrush, and Ahrefs found meaningful agreement between the three in aggregate, but real per-domain variance — meaning DA 50 on one tool is not the same underlying authority as DA 50 or DR 50 on another. Treating any specific number as an absolute threshold, rather than a comparison tool within a single provider's own scale, misreads what these metrics were built to do.
What Domain Authority Is Actually Useful For
- Comparing your own site's trajectory over time, within the same tool, rather than chasing an absolute target number.
- Sizing up competitors' relative link profiles quickly, as a rough proxy rather than a precise measurement.
- Evaluating link-building prospects — a link from a DA 10 site and a DA 70 site are not likely to carry equal weight, even though the exact multiplier isn't precisely knowable.
- Reporting relative progress to stakeholders who need a simple number, with the caveat that it's a proxy, not a ranking guarantee — a common ask when reporting on a Local SEO program's progress.
What It Isn't Useful For
- Predicting a specific page's ranking for a specific keyword — domain-level authority correlates more than page-level metrics do, but neither is a reliable individual-page predictor.
- A direct optimization target. Chasing a higher DA number through purchased links or low-quality link schemes can raise the score while doing nothing for actual rankings, and can trigger a genuine Google penalty in the process.
- Comparing across tools. A DA 50 and a DR 50 are not describing the same thing, even on the same domain.
The honest summary: Domain Authority correlates with rankings because it's built from a real signal Google does reward — not because Google looks at the score itself. Used as a comparison tool within its own limitations, it's genuinely useful. Used as a target to hit by any means, it's a metric that's trivially easy to inflate without moving the thing that actually matters — which is exactly why an AI Content & GEO approach focuses on earning genuine citations and links, not gaming a proxy score.