Two numbers, one scale, and almost everyone reads the wrong one. Domain Authority gets quoted in pitch decks and link-buying emails. Page Authority is the one that actually says something about the URL you are looking at. Both come from the same company, and neither comes from Google.
Which of the two belongs in a decision depends entirely on what you are deciding.
Key Takeaways - Page Authority predicts one URL, Domain Authority predicts a whole site. Both are Moz scores from 1 to 100, both are predictions rather than measurements
- Neither is a Google ranking factor. Moz says so plainly on its own page about the metric
- For content decisions, Page Authority is the useful one. It is per URL, so it can tell pages apart; Domain Authority gives every page the same number
- Both scores are blind to accuracy. A page with a strong link profile and a 2021 statistic scores well and still loses readers
What is Page Authority?
Page Authority is a Moz score that predicts how well one specific URL will rank in search results. It runs from 1 to 100, and it is built mainly from the links pointing at that page.
The scale matters more than people expect, because it is not linear. Moz describes it as a 100-point logarithmic scale. Moving a page from 20 to 30 is far easier than moving it from 70 to 80. A site that pushes its weakest pages from 15 to 25 does more total work than one that drags a single strong page up five points.
Two consequences follow for anyone working on content:
- Comparisons only hold inside a similar band. A 35 and a 40 are close. A 70 and a 75 are far apart.
- Movement at the bottom is normal, movement at the top is news. A new article climbing into the thirties is a crawl artefact as much as an achievement.
What is Domain Authority?
Domain Authority is the same idea applied to a whole domain. It predicts how likely a website is to rank against its competitors, on the same 1 to 100 scale. It rests on the link profile too.
Moz is direct about what the number is not:
Moz What is Domain Authority?, Moz SEO Learning Center Domain Authority is not a Google ranking factor and has no effect on the SERPs.
That sentence rules out most of the ways the metric gets used. Nobody at Google sees your DA. Raising it does not cause rankings; it reflects something that also tends to cause rankings, which is a different relationship. We went through the practical side of that in what Domain Authority actually is.
What is the real difference between them?
One number describes a page, the other describes an address. Everything else follows from that.
| Page Authority | Domain Authority | |
|---|---|---|
| Scope | One URL | The whole domain |
| Changes when | That page gains or loses links | The site’s overall link profile shifts |
| Same for every page? | No, each URL has its own | Yes, every page inherits it |
| Useful for | Comparing your own pages | Comparing your site to competitors |
| Moves in | Weeks to months | Months to years |
| Says anything about content quality? | No | No |
The row that decides most arguments is the third one. Domain Authority gives every page on your site the same score, so it cannot help you choose between two of your own pages. Page Authority can.
Both metrics are predictions of ranking ability, not measurements of it. A page can hold position three with a Page Authority of 12 if the query is narrow enough and nobody else has written a good answer. The score is a guess about the average case, and your page is a specific case.
Which one should you watch?
Match the metric to the decision rather than to the dashboard.
Watch Page Authority when the decision is about a page. Which of these two overlapping articles should become the canonical one? Which URL should the others redirect to? Which page deserves the internal links? Page-level scores answer all three, because they differ per URL.
Watch Domain Authority when the decision is about a site. Is this guest post worth writing? Is this domain worth buying? Can we realistically compete for this head term in the next year? Those questions are about the address, and a single domain-level number is the right shape of answer.
Watch neither when the decision is about your own content calendar. What to write next, what to update next and what to delete next are answered by search demand and your own performance data. A third-party prediction of link strength has nothing to add there.
Why does a page with high Page Authority still lose traffic?
Because Page Authority only knows about links, and links are the slowest-moving part of a page. Everything else can rot while the score sits still.
Three things go wrong underneath a stable score:
- The facts expire. The page keeps every backlink it ever earned and quietly becomes wrong. Nothing in the metric notices.
- The intent behind the query moves. The page still answers the 2023 version of the question.
- Better answers arrive. Competitors publish more complete pages, and relative quality drops without any link changing hands.
That gap is the most common surprise in a content audit. The pages with the best scores are often the oldest, and age is exactly what makes a page factually unreliable. A high-authority page that states a superseded number spreads that number further than a weak page ever could. We traced the mechanism in content decay.
Richard Roth Founder CiteRanger, SEO Strategist When I look at an archive, a high Page Authority on an old page makes me more worried, not less. It means that page is the one getting read, getting cited and getting pulled into AI answers. If something on it is out of date, the score is just making the damage efficient.
How do you use Page Authority in a content audit?
As one column among several, and never as the deciding one.
Where it earns its place is the pruning step. When two of your pages compete for the same query, the scores tell you which one the web has already chosen. The stronger URL is usually the one to keep and redirect into. Deleting the page with forty referring domains and keeping the one with two is the expensive mistake. A page-level score catches it before you make it. The full decision tree is in delete, merge or refresh.
A practical way to use the column:
- High score, falling traffic: update first. The page has standing it can still cash in.
- High score, no traffic: check the intent. It may rank for something you no longer care about.
- Low score, good traffic: leave it alone and feed it internal links.
- Low score, no traffic, no links: a candidate for merging or removal.
Pull the scores once, at the start of the audit, and freeze them in the sheet. Refreshing them mid-audit means your decisions were made against two different data sets, and the crawl-to-crawl wobble in these metrics is large enough to flip a close call.
What do both scores miss entirely?
Everything that is not a link. That includes the parts a reader actually judges you on.
Neither score reads a sentence. Neither checks whether the price in your comparison table is current, whether the study you cite has been replaced, whether the tool you recommend still exists. A page can hold a Page Authority of 60 and tell every visitor something that stopped being true two years ago.
For the same reason, neither score says much about whether an AI system will cite you. Systems that quote sources care about whether a claim is supported and current. Our piece on how AI search engines choose sources goes through what they do look at. The link graph is one input among many there, and a visibly stale page is a poor bet regardless of its score.
So pair the metric with something that does read the page. The content audit tool walks the URLs in your sitemap and returns the claims that no longer hold. That is the half of page quality no authority score covers.
What should you track instead?
Three things, all of which you already have access to.
Rankings and impressions per URL, from Search Console. Real position data for your actual queries beats any prediction of ranking ability. Sixteen months of it shows direction, which a snapshot score never does.
Referring domains per URL. The underlying input to both scores, without the modelling layer on top. For pruning and redirect decisions, the raw count is more honest than the compressed score.
Accuracy, as a scored column. The dimension that decays fastest and the only one with no metric in any SEO tool. Scoring it yourself, even on a rough one-to-five scale, surfaces more work than the authority column ever will.
Authority is worth building, just not worth staring at. Raising either score takes the same work that makes a site worth citing. We laid it out in raising Domain Authority without buying links.
Conclusion
Page Authority is the page-level prediction, Domain Authority is the site-level one, and both are Moz models rather than Google signals. Use the page-level score where you compare your own URLs and the domain-level score where you size up a whole site. Keep both out of decisions they were never built for.
The sharper habit is to treat a high score on an old page as a reason to check it, not a reason to relax. The pages with the most authority are the ones whose mistakes travel furthest.