What Is Domain Authority (DA)?
Domain Authority (DA) is a search-engine-ranking score developed by Moz that predicts how likely a website is to rank in search results. It runs on a 1-to-100 logarithmic scale, calculated by a machine-learning model trained against actual search results using signals such as linking root domains and total links. DA is a comparative estimate, not a Google metric.
- Domain Authority is a proprietary Moz metric on a 1–100 logarithmic scale; it predicts ranking ability rather than measuring current traffic or rankings.
- Moz calculates DA with a machine-learning model trained against real search results, using inputs including linking root domains, total links, MozRank, MozTrust, Spam Score, and anchor-text distribution.
- Moz relaunched the model as Domain Authority 2.0 in March 2019 to better track Google’s algorithm.
- Google does not use Domain Authority — John Mueller has stated Google has no equivalent site-wide authority score in its ranking system.
How Domain Authority Works
Domain Authority is Moz’s answer to a practical question SEOs ask constantly: if I go up against this domain, how likely is it to outrank me? Rather than measure a site’s current traffic or count its links, DA predicts competitive ranking ability and expresses that prediction as a number from 1 to 100. The higher the score, the more likely — in Moz’s model — that the domain ranks well across search results.
That framing matters because it makes DA a comparative tool by design. Moz is explicit that Domain Authority is best used to compare one site’s strength against another’s, or to track a single site’s trend over time, not as an absolute grade. A DA of 50 is meaningful only next to the DA of the sites you are actually competing with. In an easy niche, 50 might dominate; in a crowded one, it might not crack the first page.
Because the score is logarithmic, the effort to move it is wildly uneven. Growing from DA 20 to DA 30 is comparatively quick; pushing from DA 70 to DA 80 demands an order of magnitude more high-quality backlinks. Chasing the top of the range is a game of diminishing returns, which is why experienced practitioners watch the direction of travel rather than obsess over the exact digit.
What Goes Into the Score
Moz calculates DA with a machine-learning model, and the model is trained against real search results — it is tuned to predict which domains actually appear in Google’s rankings. The inputs Moz has described feeding into that model include:
- Linking root domains — the count of unique domains pointing to the site, the heaviest lever.
- Total links — the overall volume of backlinks, including multiple links from the same domain.
- MozRank and MozTrust — Moz’s own link-authority and link-trust scores.
- Spam Score adjustment — a downward correction for profiles that look manipulative.
- Anchor-text distribution — the mix of anchor text across the backlink profile.
Because a machine-learning model sits in the middle, there is no clean hand-computable formula. Moz periodically retrains it to stay aligned with how Google actually behaves, which means the whole scale can shift underneath you even when your own site hasn’t changed.
Example of Domain Authority
The clearest documented moment in DA’s history is the Domain Authority 2.0 relaunch, which Moz rolled out in March 2019. It is a useful worked example because Moz described exactly what changed and why, and the reasoning reveals what the metric really is.
DA 2.0 was a full retraining of the model rather than a cosmetic tweak. Moz rebuilt it to better predict rankings in the face of Google’s evolving algorithm, strengthening how the model handles manipulative link profiles through an improved Spam Score adjustment and refining its use of signals like anchor-text distribution and linking root domains. Crucially, Moz warned users ahead of the switch that individual DA scores might rise or fall on launch day — not because those sites had changed, but because the model had. That single fact is the whole lesson: DA is a prediction produced by a model, and when the model is recalibrated, the predictions move.
The 2.0 change also sharpened a point Moz has always made and that the Moz Domain Authority documentation still stresses — DA is a relative, comparative metric. After a retraining, the only honest way to read your score is against your competitors on the same day, because everyone was re-scored on the same new model at once.
And it is worth stating plainly what DA is not. It is not a Google ranking factor. Google’s John Mueller has publicly rejected the idea that Google uses a domain authority signal of this kind. DA can correlate with rankings because Moz deliberately trains it against real search results, but correlation is the mechanism, not causation. Raising DA does not push a button inside Google’s algorithm.
The single most common mistake I see is treating a DA drop as a penalty. It usually isn’t. Because DA is scored on a curve relative to the rest of the web, your score can fall even when your own links improve — simply because Moz retrained the model or because bigger sites climbed past you. I have calmed down more than one client who watched DA slide two points and assumed Google had struck them, when their traffic was flat and their rankings were fine. DA answers one narrow question: relative to other domains, how likely is this one to rank? It is a comparative gauge, not a report card, and it is not the number Google keeps. Watch the trend against your competitors, not the absolute digit, and never mistake it for a diagnosis.
Domain Authority vs Domain Rating
Domain Authority and Domain Rating are the two metrics people confuse most, because both put backlink strength on a roughly 0–100 scale. They are not the same measurement and their numbers must never be swapped.
| Domain Authority (DA) | Domain Rating (DR) | |
|---|---|---|
| Vendor | Moz | A professional SEO backlink database |
| What it predicts / measures | Likelihood of ranking in search results | Strength of the backlink profile |
| Method | Machine-learning model trained on real SERPs | Iterative link-graph score, PageRank-style |
| Scale | 1–100, logarithmic | 0–100, logarithmic |
| Google uses it? | No | No |
The distinction is real: DA is trained to predict rankings, so it folds in signals like spam and anchor text, while DR is a purer measure of link strength. A site can score high on one and lower on the other. Choose the metric your team standardizes on, and only ever compare it to itself and to direct competitors.
Frequently Asked Questions
Is Domain Authority a Google ranking factor?
What is a good Domain Authority score?
Why did my Domain Authority go down?
How is Domain Authority different from Domain Rating?
The Bottom Line
Domain Authority is Moz’s forecast, on a 1–100 scale, of how competitively a whole site could rank based on its link signals. It is a predictive, comparative model — useful for sizing up domains against one another — not a measurement of quality and not a number inside Google. Read it as a relative weather report on link strength, and judge it against your actual competitors.
Sources
- Domain Authority — Moz SEO Learning Center — Moz
- John Mueller Rebuts Idea that Google Uses Domain Authority Signal — Search Engine Journal
Roborank tracks your domain’s link authority and backlink profile over time and shows you where you stand against the sites outranking you.
Benchmark your authority →Rank & Cash — the weekly SEO breakdown
One practical teardown a week on ranking in search and getting cited by AI. No fluff.
