No, Moz Domain Authority is not a Google ranking factor. It is Moz's own estimate of a domain's relative ranking potential, built mostly from backlink data, and Google does not read that score when it ranks a page. The claim that Google uses Moz domain authority as a signal is debunked: Google has said directly that it does not use third party SEO metrics, and Moz says the same thing about its own score. A high DA can still be a useful competitive clue. It is not proof of Google trust, and it is not a guarantee that any one page will rank.
If you have ever been asked in a meeting whether raising your domain authority ranking factor score will fix your rankings, this is the piece to send back. Marketers ask "is DA real" far more often than they ask what it actually measures, and the two questions have different answers. This piece walks through what DA actually measures, why it tracks rankings anyway, and what Google and Moz have both said about it on record.
What does Moz Domain Authority actually measure?
Moz's own explainer, last updated March 18, 2025, defines Domain Authority as a search engine ranking score Moz built to predict how likely a website is to rank in search results compared with its competitors. That definition matters more than it sounds like it should, because almost every misunderstanding about DA comes from skipping past the word "predict."
A few concrete facts, straight from that explanation:
- DA runs from 1 to 100.
- A higher score means a higher predicted ranking ability, inside Moz's own model.
- DA describes a whole domain or subdomain, not the strength of one page.
- It is comparative and directional, not an absolute grade.
- Moz states plainly that DA is not a Google ranking factor and has no effect on search results by itself.
A better phrase than "Google trust" or "authority score" is predicted relative ranking potential. It is less catchy, but it is accurate, and accuracy is the whole point of a piece like this.
Under the hood, Moz's March 2025 explanation says DA runs on a machine learning model trained on data from Moz's own Link Explorer index, using dozens of factors that are mostly about backlinks: the number of linking root domains, the quality of those links, and other signals that tend to correlate with rankings across a large set of search results. A linking root domain just means a distinct site that links to you. Ten links from one site is not the same as one link each from ten different sites, and DA treats those very differently.
It helps to resist the urge to simplify DA down to a backlink counter. Moz describes a model that weighs several characteristics of a link profile together, not a running tally. And Moz has never published a full, reproducible formula with an exact weight for every input, so there is no honest way to say "link quality is worth this many points" or "domain age counts for this percentage." Nobody outside Moz has that number, because it was never published.
There is also a related metric worth a quick mention so you do not confuse the two. Page Authority, or PA, is Moz's score for predicting how a specific page will rank, defined in Moz's November 18, 2024 explainer. PA is link centric and does not account for on page elements like keyword use. Like DA, it is not an acknowledged Google ranking factor. The short version: DA predicts a domain's strength, PA predicts a page's strength, and Google treats neither one as an input.
One detail that trips people up: the DA scale is roughly logarithmic, according to Moz's March 2019 explanation from Cyrus Shepard. Moving a brand new site from a very low score up toward 10 is much easier than moving an established site from 70 to 80. A DA of 40 is not twice as strong as a DA of 20. It is also not a probability. A DA of 70 does not mean a 70 percent chance of ranking first, and it is not a percentage of Google's trust in you.
And there is no universal good score. Moz's own explainer says a DA of 30 can be excellent in one industry and mediocre in another, so the only useful comparison is against the sites actually ranking for the terms you care about. If someone hands you a DA number with no competitor context attached, they have handed you half a fact.
What evidence makes DA look like a ranking factor?
If Google does not use the score, why does it track rankings so often that people assume it must? The honest answer has three parts, and all three come from the same place: links.
Shared underlying information. Moz builds DA mostly from backlink data. Google has confirmed, in its own documentation, that it uses links as one signal when assessing a page. So a site with a strong link profile by Moz's measure can also do well in Google's, without Google ever looking at the Moz number itself. Two different graders can agree on the same homework without one copying from the other.
Prediction is the explicit design goal. Moz built DA to predict ranking outcomes, so some agreement with actual rankings is not a surprising side effect. It is the thing the model was trained to do. A model that is good at predicting an outcome is not the same as a model that causes the outcome, any more than a weather forecast causes the rain.
The arrow can point the other way. Better visibility in search can attract more links on its own, so the relationship between links and rankings is not strictly one directional. A site that is already doing well for other reasons tends to pick up links along the way.
Moz has looked at this itself. A study led by Tom Capper, published November 8, 2021, examined 10,000 keywords from Moz's own tracking set, pulling the first 20 organic results for each one, across both desktop and mobile, from a suburban US location. The analysis used Spearman's rank correlation, a method suited to comparing ranked lists rather than raw numbers. Moz reported that DA and PA outperformed a simple count of followed links as a predictor of ranking order within that sample. Restricting the analysis to just the top 10 results weakened the correlations and even changed which metric came out ahead.
The study itself is careful about what it does and does not show. It says outright that correlation does not prove causation, and that building more links does not always improve rankings. It also states, in its own words, that DA is not directly used by Google but correlates with things that are. Worth saying plainly: this is Moz's own research team evaluating Moz's own metric. That does not make the methodology wrong, but it is not independent validation, and nobody should present it that way.
It is also worth being honest about what that 2021 study cannot tell you in 2026. A five year old correlation study, from one sample of keywords, is evidence about how the metric behaved then. It is not a live measurement of how closely DA tracks rankings today, and it should never be quoted as if it were.
What do Google and Moz say about using the score?
This is where the question actually gets settled, because both sides have been asked directly and both sides answered the same way.
Google's John Mueller addressed it head on in an AskGooglebot video, reported by Search Engine Land on August 16, 2023: "No, Google does not use scores from third-party SEO tools or services. It doesn't matter whether it's about a website's authority or a rating about spamminess, Google doesn't use these scores at all." That is about as direct a denial as a company spokesperson gives. The same report adds an important nuance: those scores can still help an analyst spot real issues on a site, which is different from being a ranking input.
Google's broader guidance on third party SEO tools, last updated June 5, 2026, backs up the same point from another angle. Third party tools do not have access to Google's internal ranking data, they cannot guarantee performance, and their predictions are exactly that, predictions, which may or may not pan out. Google's own advice is to use Search Console for information that actually comes from Google Search.
Moz does not disagree. Shepard's 2019 piece asks both "is Domain Authority a ranking factor?" and "does Google use Domain Authority in its algorithm?" and answers no to both, in Moz's own words. Moz's current explainer keeps that same position: DA is a comparative metric, not a Google ranking factor. This is not a case of Google dismissing a vendor's claims while the vendor insists otherwise. Both sides describe the same thing the same way.
None of that means Google ignores links entirely, and it is just as easy to overcorrect into a different myth. Google's documentation on link best practices, updated December 10, 2025, states that Google uses links as a signal for assessing a page's relevance and for finding new pages to crawl. Google's own explanation of how it determines ranking results says links from other prominent sites can be a sign that information is trustworthy. So it is accurate to say Google cares about links. It is not accurate to jump from there to "Google's system is just DA under a different name." Google's ranking systems guide, also updated December 10, 2025, says its systems are built to work at the page level, with site wide signals and classifiers contributing as well, and it specifically warns that strong site wide signals do not mean every page on that site will rank well. Google and Moz can both look at link data without using the same index, the same weighting, or the same conclusions.
Why a high DA is not the same as Google trust
The confusion stacks three separate ideas on top of each other: a model that estimates relative ranking potential, a search engine that weighs many signals when ranking a specific page for a specific query, and a business that wants to be seen as credible. Those three ideas overlap in everyday conversation, but they are not interchangeable, and treating them as one thing is where "Google trust" talk usually goes wrong.
DA is, by Moz's own definition, a modeled estimate of link profile strength relative to other sites. It does not audit whether your facts are accurate, whether a specific page answers a specific query well, or how Google privately views your site, because no outside party has access to that. And Google's own ranking guide is explicit that strong site wide signals do not guarantee that every page wins. A domain with an excellent score can still publish a page that loses to a lower scored competitor's page, because Google is ranking that page, not the domain's reputation as a whole.
The clearest way to say this to an executive: DA is Moz's estimate of competitive ranking potential. Google trust is not something this number measures or certifies, because no such official number exists. Treat "Google trust" as shorthand your team uses internally, not as the name of a real Google score, and you will stop making promises the metric cannot back up.
It also helps to know what moves the number, so a shift in DA does not get mistaken for a shift in Google's opinion of you. Moz's explainer lists several reasons scores change: updates to Moz's own index, newly found or newly lost backlinks, changes in your competitors' scores, links that contribute little, and changes to Moz's calculation model itself. A DA drop by itself is not evidence that Google penalized you. A DA rise by itself is not evidence that Google trusts you more. The score and your actual search performance need to be checked against each other, not assumed to move together.
How should marketing teams use DA instead?
None of this makes DA useless. It just means it belongs in a specific, limited role, the same role Moz itself recommends: as one input for competitive comparison, not as a verdict on whether Google likes you.
Good uses, supported directly by Moz's own guidance: comparing your modeled link strength against the specific competitors ranking for your target terms, getting a rough sense of how competitive a given search result looks, tracking a directional trend over time alongside your other metrics, and screening potential link partners as one factor among several. Shepard's own write-up warns against using DA in isolation and recommends pairing it with actual keyword rankings, traffic, and conversions.
Where it should not make the call: do not make DA your only KPI in an executive report, do not promise rankings once you hit a target score, do not write off a relevant but lower scored site as worthless purely on its number, and do not buy a link because the seller's site has a high DA. That last one deserves its own warning. Google's policy on link spam defines buying or selling links for ranking purposes, and automated link schemes, as manipulation, regardless of what score the linking site carries. Google's link best practices guidance says paid links should be marked sponsored or with nofollow. A high DA on the seller's side does not make a manipulative link arrangement safe, and a vendor's promise to raise your score is not the same as a promise to improve your actual rankings.
One more boundary worth naming directly: nothing in the evidence here supports using DA as a stand in for AI visibility either. A high Moz score does not establish any threshold or causal link to whether ChatGPT, Perplexity, or Google's AI Overviews cite your pages. That is a different measurement question with its own evidence, and borrowing DA to answer it is exactly the kind of shortcut this piece is arguing against.
What would change this verdict
A change here would take a new, specific statement from Google saying it has started consuming a named third party score, or solid evidence of that actually happening. A single high DA site that wins a ranking, a dashboard number that moved, or an anecdote from one campaign would not be enough on its own. None of those distinguish correlation from mechanism any better than the 2021 study already did.
The useful habit to walk away with: use DA to ask sharper competitive questions, not to hand an executive a trust score Google never issued. The domain authority ranking factor question comes up again every time a client update lands, so keep the same follow up ready: what happened to the rankings, the traffic, and the conversions that were supposed to move with it?
