How to Check Any Site's Backlinks for Free (5 Methods, No Ahrefs)

Finding backlinks to a site costs nothing, provided you accept that every free source is looking at a different piece of the web. I have spent a lot of the last year inside link data, first Moz's, then a vendor API we no longer pay for, and now an index we built ourselves from Common Crawl. The thing that surprised me is how little the sources agree. The same domain routinely shows counts that differ by a large multiple between sources, and each source is telling the truth about what it crawled.

So this is less a list of tools than a guide to what each one can see. A backlink, for anyone new to this, is a link from a page on one site to a page on another. The site receiving it has the backlink. The site giving it is the referring domain. The longer version is in how backlinks work.

Google Search Console, if the site is yours

For a site you own, nothing beats Search Console. It reports the links Google itself found, which makes it the only source that answers "what does Google know" rather than "what did some crawler find".

Open the property, then the Links report. Google's documentation for it lists the sections. Top linked pages, top linking sites, top linking text, plus an internal links report. The doc says "Tables are limited to 1,000 rows, so tables may be truncated in larger or more linked-to sites." The export from the report's landing page goes further, "up to 100,000 rows of data", with a choice of "Latest links" or "More sample links". Everything is grouped by root domain, so www and bare hostnames are folded together.

The obvious gap is that it only works for your own property. No competitors, and no domain you are thinking about acquiring. When the job is evaluating an expired domain, which is most of my job, Search Console is out from the start.

Our backlink checker, built on Common Crawl

The Revised backlink checker is free, needs no account, has no usage cap, and works on any domain. Since it runs on our own index rather than a vendor's, I can be specific about what is in it.

The individual links come from the Common Crawl web archive, an open crawl of the public web, plus our own crawls. For each site linking in you get one source page carrying the link, the anchor text, the link type, and which crawl saw it. Link type is dofollow, nofollow, ugc or sponsored, or "mixed" when one site links with more than one. Headline counts for referring domains and total backlinks come from our backlink index where we hold a figure, and they carry the date that figure was last refreshed, because a count without its date reads as current when it usually is not.

One quirk I learned building it. The index stores the exact host that was linked, subdomains included. A query for wikipedia.org on its own finds nothing, because nobody links to the bare apex. en.wikipedia.org has plenty. So the tool widens every lookup to the domain and everything under it, and a result for a bare name includes links to its subdomains.

The other thing to hold in mind is that links are found in archived crawls rather than checked live. A listed link was on that page when the crawl ran. It may have gone since.

What it misses is everything Common Crawl never fetched. Common Crawl samples the web, and our index is built for expired-domain work rather than for full coverage. A thin result means the domain is not in our index. It does not mean the site has no backlinks, and it is the reason a paid tool will usually show more.

Common Crawl itself, for people who like compressed files

You can skip our tool and go to the source. Common Crawl gives its data away, and two parts of it are useful here.

The index server lets you pick a crawl, enter a URL pattern, and see which pages of a domain were captured. That settles "is this site in the crawl at all" before you go hunting for links to it.

The web graphs are the link data. Common Crawl publishes host-level and domain-level graphs regularly, each named for the three monthly crawls it spans. The domain graph, in their words, "is built by aggregating the host graph at the pay-level domain (PLD) level based on the public suffix list". There is a caveat on the same page that catches people out: "all types of links are listed, including purely 'technical' links pointing to images, JavaScript libraries, web fonts, etc." A raw edge count is therefore not a count of editorial links.

This route is free and, relative to the crawl, complete. It is also not a tool. You download edge lists and vertex files, decompress them, and query them yourself. Fine for a developer with an afternoon and a large disk. Not fine for a quick look at one domain, which is why we built the checker.

The vendor free tiers

Moz, Ahrefs and Semrush each give away a slice of their index. Their limits, from their own pages rather than from memory.

Moz's Link Explorer offers "10 free queries per month" with a free Moz Community account. The free data covers Domain Authority, Page Authority, Spam Score, inbound link count, linking domain count, anchor text and follow versus nofollow counts. Ten a month is enough to size a shortlist and not enough to work through one.

Ahrefs' free Backlink Checker reports referring domains, backlinks, Domain Rating and Ahrefs Rank, and lists a sample of top backlinks. The page quotes the index size but does not say how many rows the free report shows. In my experience it is a short list, not the profile.

Semrush's free backlink checker says "Your free backlink analysis includes a list of the website's top backlinks" alongside Authority Score and a few headline metrics, drawn from "a backlink index of more than 43 trillion backlinks and 390 million referring domains". The full tool is Backlink Analytics on a paid plan. Semrush's plan comparison describes the free tier as "For trying out the platform, with no credit card required."

The pattern across all three is the same. Their indexes are the largest, so they are the best free source for a headline count. The list is where the cap bites, so they are the worst free source for reading who actually links. I use them to size a profile and something else to read it.

The Wayback Machine, for the part the link tools skip

The Wayback Machine does not list backlinks at all. It shows what the site was, which for an expired domain is often the more important question.

Type in the domain and read the snapshot timeline. Gaps are normal. An abrupt change of subject is not. A domain that spent eight years as a regional charity and the last two as a casino affiliate carries both histories, and the links earned in the good years do not cancel the reputation earned in the bad ones.

For more than a handful of domains, the CDX server API returns the same capture list as JSON or plain text, filterable by date range and URL prefix. It is the fastest way I know to answer "when was this last a real site" across a list.

Which one for which job

Pick two. For a domain I am evaluating I run our checker for the list, one vendor tier for the headline count, and the Wayback Machine for the history. That is a few minutes a domain once you are used to it.

Reading what comes back

Whatever the source, the same handful of numbers come back, and they get misread in the same ways.

Backlinks counts every link. Referring domains counts every distinct site. One site can put the same footer link on ten thousand pages, which is ten thousand backlinks and one referring domain. Read referring domains first, always. A hundred links from fifty domains beats a hundred from five.

Dofollow share is the next thing I look at. Google's guidance on qualifying outbound links asks sites to "Mark links that are advertisements or paid placements (commonly called paid links) with the sponsored value", to mark "user-generated content (UGC) links, such as comments and forum posts, with the ugc value", and to "Use the nofollow value when other values don't apply". Then it says "Links marked with these rel attributes will generally not be followed." A profile that is 90 percent nofollow from forums and comment sections is a profile Google has been told to discount. Mostly dofollow from editorial pages is the opposite, and rarer.

Anchor text, the clickable words, tells you how the link got there. A natural profile is mostly the brand name, the bare URL, and phrases like "this study" or "here". When a third of the anchors are the same commercial phrase, somebody built that profile rather than earned it. Google's spam policies define link spam as "the practice of creating links to or from a site primarily for the purpose of manipulating search rankings", and exact-match anchors at scale are the most visible tell there is.

Then the step every tool skips, and the one that matters most. Read the referring domains as a list of names. Open three or four of the linking pages. Decide whether a person put the link there or a script did. The order I do this in, and the checks that have saved me from bad domains, are in how to evaluate an expired domain's link history.

If you want one number to sort a long list before reading any of it, the domain authority checker gives a 0 to 100 score computed from the domain's position in the Common Crawl graph, with both underlying ranks shown next to it. It is a sorting aid. It is not a verdict, and how it differs from Moz's and Ahrefs' scores is the subject of Domain Authority vs Domain Rating vs Revised Authority.

FAQ

How do I find backlinks to my site for free? Verify the site in Google Search Console and open the Links report. It lists the linking sites, linked pages and anchor text Google found, up to 1,000 rows per table and 100,000 in the export. For a second opinion from a different crawler, run the domain through a free checker.

How do I find competitors' backlinks for free? Search Console only covers sites you own, so use a checker that works on any domain. The Revised backlink checker has no signup or cap and lists source pages and anchors. The Moz, Ahrefs and Semrush free tiers give larger headline counts but cap the list.

Is there a completely free backlink checker with no limits? Common Crawl's published web graphs have no limit and no fee, but you query the raw files yourself. Among hosted tools, ours has no usage cap and no account, with a per-domain row cap on the list. The vendor free tiers cap queries per month or rows per report.

How do I check backlink quality? Read referring domains rather than backlinks, check the dofollow share, look at the anchor distribution for repeated commercial phrases, and open a few of the linking pages to see whether the link sits in editorial prose or a footer. Then check the domain's archived history for a change of subject.

Does Google Search Console show all my backlinks? No. It shows a sample of what Google found, truncated at 1,000 rows per table. Google's own documentation says larger sites will see truncated tables. The export gives up to 100,000 rows.

Once you can read a link profile, the Revised directory of expired domains with backlinks is where the skill pays. Every listing shows the referring-domain band, the sources linking in and the archived history before you reveal the name, and the Wikipedia and Reddit source hubs let you start from referring domains you already trust.