Free Broken Link & Link Profile Checker
This does not show you backlinks, referring domains or a domain authority score, because none of that can be read from your own website. What it does show is the half of link health you publish and control: broken links, redirect chains, internal linking, anchor text and where the page sends people next.
What this checks
Broken destinations: a sample of the page's links requested directly, with the failing URL and its status code named rather than a count you have to go and find.
Links with nothing in them: anchors containing no text and no image, which a crawler reads as a link about nothing and a screen reader announces as a raw URL.
Redirect hops links that do not land where they point, which is usually old http:// or non-www addresses left in the content after a migration.
Internal linking: how many routes the page offers into the rest of the site, and whether it is effectively an orphan with nothing but a menu.
Anchor text links whose text is "click here", "read more", or "here" are each one a ranking signal you controlled outright and threw away.
Outbound links: how many other websites the page sends people to, how many distinct domains that is, and which ones come up most often.
Nofollow use: whether no-follow, sponsored, and ugc are applied selectively or blanket-applied to everything, which tells Google you vouch for nothing.
Insecure and unsafe links: plain http:// links published from an HTTPS page, and new-tab links missing rel="noopener".
Link density links per hundred words, which separates a page with something to say from a page that is really a list of links.
The SEO around the link title, meta description, canonical, heading structure, and whether anything on the page is telling search engines to stay away.
This cannot see your backlinks, and nor can anything that reads your own site
Say it once, plainly, at the top. This tool has no backlink index. It cannot show you inbound links, referring domains, domain authority, or a spam score, and it never will. That data only exists inside commercial crawl indexes — Ahrefs, Majestic, Semrush and a handful of others — built by companies spending serious money crawling the web continuously for years. Nothing that reads your own website can produce it, because who links to you is not written anywhere on your site.
Plenty of free tools present a number anyway. Where it is not simply invented, it is usually a proxy assembled from whatever the tool could reach, dressed up with a gauge and a colour. A made-up authority score is worse than no score, because people make decisions on it.
What is left when you take that away is not a consolation prize. Your outbound and internal linking is the half of link health you own outright — you can change it this afternoon, for nothing, without asking anyone — and on most sites it is the half nobody has ever looked at. Backlinks take years and other people's goodwill. This takes an editor and an hour.
What internal linking actually does
Internal links do two jobs at once, and most sites do neither on purpose. The first is routing: a crawler arriving on a page can only reach what that page links to. The links are the road network. A page with no links out of it is a dead end, and a page nothing links to is an orphan — it can sit in your sitemap for years, technically published, essentially invisible, because nothing on the site ever points at it.
The second is distribution. Whatever standing a page has acquired flows along its links to the pages it points at. Your homepage is usually the strongest page you own; if it links only to the menu items, that is where all of it goes. Link from the homepage body and from your best article to the service page you want to rank, and you have moved something of value to where it earns. That is not a trick; it is how the web has been read since the beginning.
A navigation menu is not internal linking. It is identical on every page, so it tells a search engine nothing about what this page in particular is about, and its wording is fixed by the design. Real internal linking is contextual — a sentence in the body text that names the thing and links to it, written the way somebody would search for it. Find the pages that already get traffic, and link from them to the pages that need it.
Broken links and what they cost
A dead link costs you three different things. A visitor clicks it and hits a 404, and a 404 in the middle of a buying journey is not a minor inconvenience — it is the moment someone who was reading about your service goes back to the search results, where your competitors are. A crawler follows it and spends part of its visit on a page that does not exist, which on a large site means real pages get crawled less often. And if the link had accumulated any value, that value now goes nowhere at all.
They appear on their own. A page gets renamed, a supplier changes their site structure, a PDF gets tidied out of the media library, a plugin that built a directory gets deactivated. Nobody breaks a link on purpose, which is exactly why nobody notices — you do not click your own links, and the people who do find them do not tell you.
The fix depends on what broke. If the destination has moved, redirect the old address to the new one with Redirection or your SEO plugin, rather than deleting it — anyone who bookmarked or linked to the old URL then still lands somewhere useful. If the destination is genuinely gone, edit or remove the link; leaving it and relying on a pretty 404 page is not a fix. And where the broken link is to somebody else's site, either find where the page moved to or drop it.
Anchor text, and linking out like you mean it
Anchor text is one of the few ranking signals you hand over yourself, in your own words, with nobody to negotiate with. Every time a link reads "click here", you have described the destination as "click here". Every "read more" describes it as "read more". Ten of those on a page is ten signals thrown in the bin, and it is the single easiest thing on this list to fix: rewrite the link to name what is on the other end — "our kitchen fitting service" rather than "read more", "the Hartley Road case study" rather than "learn more".
Do it naturally. Descriptive is the goal, not repetitive: forty links across the site all reading "emergency plumber Leeds" is a pattern, and patterns get discounted. Write the sentence first and let the link fall where the words already are. The same change makes the page better for anyone using a screen reader, who navigates by pulling up a list of every link on the page — a list of nine identical "read more" entries is unusable, and a list of named destinations is a table of contents.
Linking out is worth doing too, and most business sites are frightened of it for no reason. Linking to a trade body, a manufacturer, a standard or a source is what a real business does; it costs you nothing, and hoarding links does not concentrate authority — it just tells nobody anything. Keep nofollow for paid placements, user-submitted content, and links you genuinely do not vouch for; blanket-nofollowing every outbound link is cargo-cult SEO. Point commercial pages at your own contact route rather than at somebody else's site, keep social icons in the footer where they are not the first thing a bored visitor clicks, and set rel="noopener" on anything opening in a new tab.
What this link checker cannot see
Two limits matter, and both are about scope. The first is the one above: no backlink index, so nothing here describes who links to you. The second is that this reads one page, not a site. It parses every link on the page you gave it, and then requests a small sample of them — a handful, internal links first, because those are the ones you can actually fix — to see whether they answer. A page carrying a hundred links gets a handful tested, and the report says so rather than implying the page is clean.
That is a deliberate choice about being a polite visitor to somebody else's server rather than a limitation of the idea. But it means this is a spot check, not a broken link audit. For the whole site, you need a crawler: Screaming Frog will do up to five hundred URLs free on your desktop and is the standard answer; the Broken Link Checker plugin will monitor a WordPress site continuously from inside it, and Google Search Console will show you the 404s Google has already found on pages it has crawled.
Beyond that, the usual caveats apply. The page is read as it is served, so links that only appear after JavaScript runs are invisible here — which is worth knowing, because they are partly invisible to search engines as well. A server that refuses our HEAD request is reported as untested rather than broken, because calling a working link dead would be wrong. And no tool can tell you whether the page you linked to is any good; that remains a judgement, and it stays yours.
Questions people ask
Can this check my backlinks?
No. It has no backlink index and cannot show inbound links, referring domains, domain authority or spam scores. That data only exists inside commercial crawl indexes, and it cannot be derived from your own website, because your site does not know who links to it.
The free place to start is Google Search Console. Its Links report comes from Google's own crawl of the web, it is free for any site you can verify, and it shows your top linking sites and your most linked pages. For a proper backlink audit — disavow decisions, competitor gap analysis, historical link data — you need a paid tool such as Ahrefs, Majestic, or Semrush, and there is no free substitute worth relying on.
It only tested a few links. How do I check the whole site?
This requests a sample from the one page you gave it, so it is a spot check. For every link on every page, you need a crawler. Screaming Frog is the standard tool and its free tier covers five hundred URLs, which is most small business sites. The Broken Link Checker plugin runs inside WordPress and keeps watching after you stop looking, which suits a site with a blog that keeps growing.
Google Search Console is the third piece, and it is free: its Pages report lists the 404s Google has already found, which is the subset that is actually costing you something in search.
What is an orphan page?
A published page that nothing else on the site links to. It usually happens after a redesign, when a page is dropped from the menu but never deleted, or when a landing page is built for a campaign and linked only from the advert. It exists, it loads, it may even sit in your XML sitemap — and a crawler moving through your site by following links never arrives at it.
A sitemap entry helps a page get discovered but it is a weak signal compared with being linked to from a real page. If the page matters, link to it from somewhere relevant that already gets traffic. If it does not matter, delete it and redirect it to whatever replaced it.
How many internal links should a page have?
There is no target number worth chasing. What matters is that the links are relevant and that the important pages are reachable in a couple of clicks from wherever someone lands. The report flags a page as a dead end when it publishes almost none, because a page linking nowhere passes nothing on to the rest of the site.
The other end is real too. Once a page is mostly links and hardly any text — this measures links per hundred words — it reads as a navigation or archive page rather than a page with something to say, and those are very hard to rank for anything.
Should I nofollow my outbound links?
Not as a rule. Nofollow is for links you do not vouch for: paid placements and adverts, which should be rel="sponsored"; user-submitted content such as comments and forum posts, which should be rel="ugc"; and anything you are citing as an example without endorsing. An ordinary editorial link to a supplier, a standard, or a source should be followed.
Not following everything does not preserve anything. It tells Google you stand behind none of your own references, and if the report says every single outbound link on the page is nofollowed, the cause is almost always a plugin setting doing it automatically rather than a decision anyone made.
Is this link checker free?
Yes, on any page you like, within a fair-use limit that stops it being used as a bulk scanning service. There is no account to create and nothing to install.
Do I need to give you my email address?
The score, the breakdown and the first few findings are shown to everyone. The rest of the fix list asks for an email address, because that is how a free tool pays for itself. You are never asked twice.
Does running this change anything on my site?
No. It reads the page the way a visitor does, then sends a small number of HEAD requests to see whether the links resolve — the lightest request there is, asking only for the status of the destination rather than downloading it. Nothing is submitted, nothing is written, and nothing on your site or on the sites you link to is altered.
Want the whole site crawled, not one page?
We crawl every page, find every dead link and redirect chain, and rebuild the internal linking so your best pages are the ones the rest of the site points to. Send us the address, and we will tell you what a full crawl turns up.
