How to Check Backlinks (and What to Do Next)

Three different tools can look at the same website and report three different backlink counts, and none of them are lying.

Checking your backlinks means pulling the list of pages that link to yours from a source you trust, then reading what's actually on it: who's linking, how, and whether it's doing anything for you. Google Search Console does this for free, for your own verified site only. A handful of independent crawlers do it for any domain, at a fraction of what Google itself has already indexed.

The number most people chase is the wrong one.

Total backlink count is the easiest metric to inflate and the least useful one to act on. What matters is where the links come from, how many separate domains they're spread across, and whether the one bad link you just found is worth touching at all. Here's how to pull the real list, read it the way it's actually meant to be read, and decide what to do about it this week.

What "Checking Your Backlinks" Actually Means

"Checking backlinks" sounds like one action. It's actually two questions stacked on top of each other, and most guides only answer the first.

Question one: who links to my site, according to this particular tool? Question two, the one that decides what you do next: how close is that list to complete, and does the gap actually matter?

No crawler sees the whole web. Ahrefs runs its own independent crawler around the clock, and its live index currently carries roughly 493 billion pages, 209 million domains, and 35 trillion external backlink records (ahrefs.com). Moz runs a separate crawl entirely, putting its own index at over 45 trillion links across roughly 1 billion domains (moz.com). Both are enormous. Neither is Google's index, and Google has never published the size of its own.

Search Console is different in kind, not just size. It only shows links Google's own crawler has already found and stored against your verified property, and Google's own documentation is upfront about the gaps: the report includes links found "over time," some of which may since have been removed or the page deleted, and it doesn't flag whether a link carries a nofollow attribute at all (support.google.com).

Search Console costs nothing because you're not buying access to a crawl. You're looking at data Google already collected while it was indexing your own pages anyway. A third-party checker sells you access to a separate crawl it built and maintains on its own, which is why every free tier caps out fast: crawling that much of the web isn't free for the company selling it to you either.

So three tools can hand you three different numbers for the same domain, and all three can be telling the truth about what they, personally, have found. That's not a reason to distrust the data. It's the reason nobody experienced treats a single backlink count as a score to hit.

The failure mode is reacting to whichever number shows up first: panicking because a paid tool comes back thin, or celebrating because a free one shows a hundred rows. Both reactions skip the real question, which was never "how many." It's "from where, and is any of it moving anything." Buying traffic outright is a different question with its own math (see what a $0.001 click actually buys); backlinks are the slower, unpaid version of the same goal, getting a stranger from somewhere else to click through to you.

Check your own site for free in Search Console first. It's the closest thing to Google's own view of your links, and it costs nothing. Bring in a third-party crawler only when you need to see someone else's backlinks too, since Search Console only shows the property you've verified.

How to Check Your Own Backlinks in Google Search Console (Free)

If you haven't already, verify your site in Search Console. Go to search.google.com/search-console, choose "Add property," and verify either the whole domain through a DNS record or a single URL prefix through an HTML tag or your analytics account. That's the only setup cost involved, and it takes a few minutes.

1. Open the property, then click Links in the left sidebar.
2. Under "External links," open Top linking sites. This shows the root domains linking to you most, sorted by link count, with subdomains folded into their root domain.
3. Open Top linking text to see the anchor text other sites use when they link to you, and Top linked pages to see which of your own pages earned the most links.
4. Click into any linking site to drill down to the exact pages doing the linking, and which pages on your site they point to.
5. Use the export option if you'd rather work through the full list in a spreadsheet than the paginated report view.

Here's the trap most people fall into at step 2. Say your homepage shows 40 external links total, but sorting by top linking sites shows only 6 root domains. That's not 40 relationships. It's 6, and if one of them accounts for 30 of the links (a syndicated widget, a footer credit, a site that mirrors your feed automatically), you effectively have 5 real relationships and one repeat visitor wearing a big number.

Check whether the property covers what you think it covers, too. A domain-level property (verified by DNS) rolls up www, non-www, and every subdomain into one report. A URL-prefix property only counts links to that exact prefix, and if you set one up years ago for https://www and your site now serves https://non-www, you're reading half your own link profile without realizing it.

The report has real gaps. It doesn't separate follow from nofollow links, so a comment-section mention and an editorial citation look identical in the table. It only reflects what Google's crawler has already found and recorded, so a link that went live last week may simply not be there yet. And it only covers your own verified property. If you want to see what's pointing at a competitor, you need a different tool.

Do this the same week: pull the export, sort by domain rather than by raw link count, and count how many of your "backlinks" are actually the same 3 or 4 sites talking to you more than once. That number, not the total, is your real starting point.

Free and Paid Backlink Checkers, Compared

Search Console only covers your own site. If you want to check a competitor's backlinks, or want a second read on your own that includes authority scores and follow or nofollow flags, you need one of the independent checkers.

Run the same domain through two of these and the counts won't match, and that's expected given the index sizes above, not a sign either tool is broken.

Here's a worked comparison. Say you run a mid-sized blog through both the Ahrefs free checker and Moz's Link Explorer. Ahrefs might report a Domain Rating of 42 based on its own crawl. Moz might report a Domain Authority of 38 for that same site. These aren't measuring the same underlying math. A DR of 42 doesn't mean the site outranks a DA-40 site measured somewhere else. Each score is only comparable to other scores from that same tool. Comparing a DR to a DA is comparing a ruler to a scale.

What you actually want from a paid tier, if you go there, is the stuff the free versions cut: the full backlink list instead of the top 100, the historical growth-and-decline graph, and lost-link alerts that flag when a domain that used to link to you has taken the link down. Some tools also estimate the organic search traffic each linking page receives, which is a proxy for how much that mention is worth, not a promise about what it sends you directly; for what counts as a strong number once a visitor actually lands on a page, see CTR meaning and what counts as good.

If your site is small enough that Search Console's free view already shows every real domain linking to you, you probably don't need a paid tool yet. Check the free tier's row count against Search Console's own list first. If they roughly agree, that's your answer, and it's the cheapest one available.

What the Numbers Actually Tell You

Two numbers get reported everywhere and they measure different things. Backlinks are every individual link, however many times the same site repeats it. Referring domains are the count of unique sites doing the linking. If one domain links to you 30 times through a repeated widget, that's 30 backlinks and 1 referring domain, and the referring domain figure is the one worth building a strategy around, because a link from a brand-new domain reads as a stronger, rarer signal than a 31st link from a site that's already vouched for you.

Domain Rating, Domain Authority, and Authority Score are each proprietary, tool-specific estimates of how much a site's own backlink profile is likely to help pages rank. None of them are Google metrics; Google has never published anything equivalent. Use them to compare within one tool, never across tools, and never as a target in themselves. A score of 60 built on a single spun-up network of low-quality sites is not a better outcome than a score of 30 built from ordinary press mentions, even though the number looks bigger.

Follow versus nofollow used to be the whole conversation. It still matters, but less than it did, because Google's spam policies have moved from punishing individual link attributes toward pattern detection across a whole profile. The guidance, updated as recently as August 28, 2026, defines link spam as links created "primarily for the purpose of manipulating search rankings," a description of intent and pattern, not a specific HTML attribute you either have or don't (developers.google.com).

Anchor text is the one most people ignore until it becomes a problem. If every link pointing at your site uses the exact same commercial phrase as the clickable text, that pattern reads as manufactured whether it actually was or not. A natural link profile has a messy mix: your brand name, your bare URL, "this tool," "read more," and only occasionally the exact phrase you'd want to rank for. Pull the anchor text report from whichever tool you're using and look for one phrase dominating more than a third of your links. That's the number that should make you pause, not the total link count.

Run this checklist against your own export before deciding anything is actually wrong:

• More than 3 to 5 unique referring domains for any page you actually want to rank
• No single domain responsible for more than half your links to that page
• No single anchor phrase used on more than a third of your links
• At least some links from domains that plausibly know your business, not link directories with no other content

Fail one of these and you have something worth investigating. Fail none of them and the backlink profile isn't your problem, whatever else is going on with that page.

Where to Get More (Real) Backlinks Once You've Checked

Checking is diagnostic. It tells you what you already have. Getting more is a separate job, and the fastest route most small brands reach for first, asking a friend for a reciprocal link, is exactly the pattern Google's own policies flag: a coordinated link exchange, where the mutual link is the point rather than a side effect of a real relationship, sits in the same category as a paid link with no rel="sponsored" tag (developers.google.com).

What actually earns a link is something a site would cite even if you never asked for it. A comparison table nobody else built honestly. A number from your own business that a journalist can quote without checking three other sources first. A free tool that solves one specific problem well enough that whoever's writing about that problem links to it instead of describing it from memory. All three take longer to build than an email asking for a swap, and all three keep working after they're built.

If you're starting from close to zero, a curated list of directories and startup listing sites is still the fastest handful of links most small brands can pick up in a single afternoon, and The Board's list of free places to submit your backlink rounds up the ones that are actually still live and actually still pass a real link, not a nofollowed mention buried in a graveyard directory. More on the rest of what fills in around that first push lives in how to promote a new website in its first 90 days.

One source is worth naming directly, because it's disclosed rather than buried in a roundup. The Board is a public leaderboard where a brand's rank is sorted purely by what's been paid, and every listing carries a real outbound link to the brand's own site, disclosed as a paid placement every time it's shown. That's a backlink from a domain with actual visitors, not a link farm, and you can check it the same way you'd check any other link, in the same Search Console report covered above.

New backlinks take weeks to show up anywhere, whichever source they came from. Google has to crawl the linking page, then crawl or re-crawl yours, before either Search Console or a third-party tool reflects it. Don't check daily. Check monthly, and compare the export against last month's, because the number worth watching isn't the total. It's what's new and what's disappeared, and whether the traffic a link sends is worth more than what an equivalent visitor costs to buy outright.

When a Backlink Is Actually a Problem (and When to Ignore It)

Most backlinks that look bad aren't worth touching. Google's own John Mueller said as much directly, and recently:

If you're conflicted and just want to be sure, it's totally fine to set up and use disavow files. If you notice that the bulk of the problems are from a few TLDs, you can also disavow the whole TLD. The disavow file is a tool, not a religion. Most sites don't need it, but that's not all sites.

That's from March 2026, and it's the clearest official answer to a question small site owners ask constantly (Search Engine Roundtable, seroundtable.com, March 6, 2026).

Save the concern for a pattern, not a single link. A spammy comment or a scraper site that copied your page wholesale is exactly the kind of link Google's own systems are built to discount without any help from you. What's actually worth watching for is different: a sudden spike of dozens of new low-quality domains linking with the same commercial anchor text in the same week, especially if you didn't do anything to cause it.

If you do see that spike, work through it in order. Confirm it's real in at least two tools, since a single tool glitching is more common than a real attack. Check whether the new linking domains share an obvious pattern, the same registrar, the same template, the same posting date. Only then consider a disavow file, built at the domain level rather than link by link, since Mueller's own guidance points to a domain-wide disavow as the more efficient move when the bulk of the problem sits on a handful of TLDs.

For everything else, keep checking on a monthly cadence and keep the referring-domain count and the anchor text mix as the two numbers you actually watch. A site with 12 referring domains, half of them earned in the last quarter, is in a stronger position than a site with 200 backlinks from the same eight partners it's had since launch.

That's the actual measure of a healthy backlink profile: not the total, and not any single tool's score for it, but whether the list of who's linking to you keeps growing with names that weren't there the last time you looked. More on what decides whether a visitor sticks around once they've clicked through lives under Traffic.