DA vs DR: Domain Authority vs Domain Rating

DA and DR are not two names for the same number. Domain Authority (DA) is Moz’s score: a machine-learning model, fed by Moz’s link index, that estimates how likely a domain is to rank in search results. Domain Rating (DR) is Ahrefs’ score: a measure of how strong a domain’s backlink profile is, computed from the link graph in Ahrefs’ index. Both print a number on a 0–100-style scale — and that shared scale is exactly where the confusion starts, because the data and maths behind them are entirely different.

The mix-up costs buyers money. Link packages are sold against floors like “DR 40+” or “DA 50+”, and a buyer who treats the scores as interchangeable turns down strong sites because the wrong number looked low — or pays a premium for a domain whose impressive score came from a tool they do not even use. Below: how each is built, why the same website can read 62 in one and 41 in the other, and how to use each when you are paying.

The one-line difference: whose crawler, whose model

If you remember one sentence, make it this one. DR answers “how strong is this domain’s backlink profile, according to Ahrefs’ crawl of the web?” DA answers “how likely is this domain to rank, according to a model Moz trained on ranking behaviour?” One reads a link graph; the other predicts an outcome — and every buying mistake in this topic follows from confusing the two.

How each score is built

DR: the strength of a domain’s own backlink profile, on a 0–100 scale

Ahrefs keeps DR narrow: it reads the backlink graph and nothing else. Three ingredients decide the number. First, how many domains link to you — referring domains, not raw link counts, so a thousand links from one site weigh far less than a hundred links from a hundred sites. Second, the strength of those linking domains: a followed link from a DR 70 domain contributes more than one from a DR 10 domain. Third, dilution: a domain that links out to hundreds of sites passes a thinner slice of its strength to each, so a link from a generous directory is worth less than the same link from a site that links out sparingly.

Only followed links count, and the scale is relative — your DR is positioned against every other domain in Ahrefs’ database, so it can drift down in a period when you lost nothing, simply because neighbouring domains gained.

DA: a ranking-prediction model, also 0–100

Moz takes the opposite decision. DA is the output of a machine-learning model trained to predict how likely a domain is to appear in search results. Its inputs are link-based — linking root domains, total links, and Moz’s own intermediate link metrics — but there is no fixed formula: the model sets the weightings, fitted to observed ranking behaviour rather than to the link graph alone. That is why DA sometimes rates a domain generously whose raw link profile looks ordinary: if domains with that profile demonstrably rank, the model says so.

The trade-off is movement in blocks: when Moz refreshes its index or retrains the model, scores shift across the whole web at once, and your DA can drop in a month when you changed nothing. Both scales also get harder as you climb — moving from 20 to 30 takes a fraction of the link strength needed to move from 70 to 80 — so a small gap at the top means more than a small gap at the bottom.

Why the same site scores 62 in one and 41 in the other

Take a hypothetical mid-sized B2B blog: Moz shows DA 62, Ahrefs shows DR 41 (illustrative numbers, a pattern we see constantly). Neither tool is broken — work through the usual causes and the gap explains itself:

• Different indexes. Moz and Ahrefs crawl the web independently, and each finds links the other has missed — two databases, two versions of the same site’s profile.
• Different counting rules. DR counts followed links only. If a good share of this blog’s profile is nofollow — profiles, comment threads, social and directory entries — that share feeds Moz’s link data while adding little or nothing to DR.
• Dilution. DR divides a linking domain’s strength among all the sites it follows out to. Links from directories and resource pages that link to thousands of domains pass very little each in Ahrefs’ maths.
• Model versus graph. DA’s model can lift a domain that ranks well for its niche even with a modest graph, because ranking is what it was trained to predict. DR has no opinion about rankings — it reads links, full stop.
• Timing. The two companies recalculate on different schedules, so scores checked on the same morning are still snapshots from different moments of two different indexes.

None of this makes one score “the true one” — they are two estimates built from two datasets for two purposes. The error is treating them as the same currency and converting between them as if an exchange rate existed.

Which score matters for which buying decision

Start from the limit both scores share: no search engine uses DA or DR. They are third-party estimates — useful as filters when sorting candidate domains, meaningless as promises about what a link will do. Which one leads a purchase depends on one practical question: which tool does your own reporting run on? The floor you shop by should be the score you can verify on the day the work is delivered.

Shopping by DR: what the floor actually filters

A DR floor filters for link-graph strength in Ahrefs’ index. A domain clears a DR 40 floor because other strong domains chose to link to it with followed links — which filters out thin, new or self-made profiles. Be clear about what the floor does not filter: relevance (a DR 50 gambling blog passes a DR filter for a plumbing client and is still the wrong link), organic traffic, editorial quality, or the strength of the specific page your link lands on. The floor is a first sieve, never the whole inspection.

It also has to be auditable. Our own DR package is built to that standard — links on DR 40+ domains, with each domain’s DR recorded at delivery — because a floor you cannot re-check in Ahrefs afterwards is just a claim in a sales title. Before ordering from anyone, check the report format you will receive; our sample report shows the metric printed beside each live URL.

One standing honesty note: links bought to influence rankings sit inside Google’s link-spam policy area, and no score makes a purchased link safe, policy-proof or guaranteed to move anything. Scores help you choose placements; they do not change what the placement is.

Shopping by DA: the same logic in Moz’s index

If your stack is Moz — tracking, client reports and your site’s history all run on DA — shop and verify in DA, and let DR be an occasional cross-check. A DA floor filters for domains Moz’s model rates as likely rankers, loosely folding some ranking behaviour into the screen alongside the link data. The evidence rule is identical: the delivery report should print each domain’s DA, recorded at delivery and checkable in Moz — the standard we argue for in our guide to what a good DA for backlinks actually looks like.

What you must never accept is one score offered as proof of the other: a seller promising “DR 40+” who delivers domains checked only in Moz has not delivered what was sold. Sellers know buyers mix the two up, which is why the higher number is always the one in the pitch. Ask which tool the floor was measured in — and check it there.

Using both without fooling yourself

Five working rules keep the two scores honest:

1. Never average them, and never convert. There is no exchange rate between DA and DR. A rule like “DR runs about ten points below DA” invents a precision neither vendor claims, and misleads you most on the unusual domains where the gap is the information.
2. Record tool, score and date together. Both scores drift as indexes refresh, and an undated “DR 45” in an old spreadsheet may describe a domain that has since changed hands or lost its best links. A score without its source and date is not evidence.
3. Read the gap as a question, not an answer. High DA with low DR means the domain ranks better than its followed link graph suggests. Low DA with high DR means a strong graph with a weaker ranking footprint — worth a manual look at where those links come from before you pay, because a graph-only score can be lifted by graph-only tactics.
4. Remember both scores describe domains, not pages. Your link sits on one page, and each tool scores pages separately — PA in Moz, UR in Ahrefs. The gap between the two numbers is large enough that we gave it its own guide on domain authority vs page authority.
5. A score says nothing about indexing. A link on a DR 60 domain passes nothing while the page carrying it sits outside Google’s index. Before debating scores, check whether the page carrying your link is indexed.

That is the whole method: DR is your link-graph ruler inside Ahrefs’ data, DA is Moz’s ranking-likelihood estimate inside Moz’s data, and each is trustworthy exactly as far as its own definition reaches. The mistake was believing the numbers were translations of each other.

Frequently asked questions

Is a DR 40 site equal to a DA 40 site?

No. The two 40s come from different data, calculations and definitions, so any resemblance is coincidence. Both sit mid-range in their own scale, but sites can only be compared meaningfully inside one tool at a time.

Why is my DR much lower than my DA?

Usually a combination of ordinary causes: much of your profile may be nofollow (DR counts followed links only), your links may come from domains that link out to many sites (diluting DR), Ahrefs’ index may hold fewer of your links than Moz’s, and DA’s model credits ranking behaviour DR ignores. A gap on its own is not a defect to fix.

Which is more accurate, DA or DR?

Neither measures a real Google quantity, so “accurate” has nothing to be accurate against. Each is internally consistent inside its own system. Use the one your tools and reporting already run on, and cross-check in the other before spending serious money on a domain.

Do any search engines use DA or DR?

No. Both are third-party estimates; Google runs its own link analysis and does not consume either score. They correlate with rankings because they summarise link strength, which search engines do care about — but correlation is not an input.

Should I track both scores for my site?

You can, provided you track each against its own history, never against each other. For most sites, one primary score — the one your reporting uses — plus a cross-check when the gap suddenly changes is plenty. Watching both weekly mostly produces two numbers to worry about instead of one.

If DR is the score your reporting runs on, our High DR package places 75 manual links on DR 40+ domains and prints the DR beside every URL in your report. If your stack is Moz, the same evidence rule applies to every DA package in the store: pick the score you can verify, and hold every seller — us included — to printing it next to the link it describes.