Free Canonical Tag Checker

What a canonical tag tells Google, and when it points at the wrong page.

One line in the head decides which of a page's many addresses collects its links and its rankings. Here is how the tag works, where it is only a suggestion, and the mistakes that hand your ranking to a copy.

What the tag is

A canonical tag is a link element in a page's <head> that names the preferred address for that page's content. Search engines use it when the same content, or nearly the same content, can be reached at more than one URL. Instead of treating each address as a separate page competing with the others, Google can fold them into one and show that one in results. The syntax is a single line:

<!-- inside <head>, on every version of the page -->
<link rel="canonical" href="https://yourdomain.com/services/estate-planning/">

Three details matter. The tag only counts in the <head>; Google's documentation says a canonical link is only accepted there, so a theme that prints it in the body has no canonical at all. The href should be a full, absolute URL with the scheme and host; Google supports relative paths but recommends absolute ones because relative paths cause problems over time. And for files with no HTML head, such as PDFs, the same instruction can be sent as an HTTP response header, Link: <https://yourdomain.com/guide.pdf>; rel="canonical". This checker reads the HTML tag only, so a header canonical will show as missing here.

Why one page ends up with many URLs

Most sites create duplicate addresses without meaning to. None of these is a problem on its own, and Google says some duplicate content on a site is normal and not a spam violation. The cost is that links, shares and ranking signals get split across addresses unless something tells Google which one is the real one.

• Tracking parameters. A newsletter link adds ?utm_source=newsletter, and now the article exists at two addresses with identical content. One of the saved runs above is exactly this case on Willowdale Equity.
• Trailing slashes. /blog and /blog/ are different URLs to a crawler. Most servers redirect one to the other; many serve both.
• http and https, www and bare. Four possible hosts for every page. A site that has moved to https but still answers on http has a duplicate of every page it owns.
• Sorting and filtering. Listing pages that take ?sort= or ?page= parameters can generate hundreds of addresses from one template.
• Print and AMP versions, preview links from a CMS, and the same service page published once per office location.

The self-referencing canonical

A self-referencing canonical is a page whose canonical tag names its own address. It sounds redundant, and it is the most useful form of the tag. Every duplicate of the page, the parameter version, the trailing-slash version, the scraped copy that kept your markup, now carries a tag pointing back to the clean URL. Google's documentation recommends adding the same self-referential canonical to the canonical page itself, and most WordPress SEO plugins add one to every post by default. It is recommended practice rather than a requirement: a page with no canonical is not penalised, it just leaves the choice entirely to Google.

When this checker compares the tag with the page, it compares against the address the fetch landed on after redirects, ignoring letter case, a trailing slash and any #fragment. So https://yourdomain.com/about and https://yourdomain.com/about/ count as the same page. A query string does not get the same allowance: a URL carrying ?utm_source= whose tag names the clean address is reported as pointing elsewhere, which is exactly what it should be doing.

A canonical URL checker cannot see Google's choice

The tag is a request, not an order. Google's own wording is that indicating a canonical preference is a hint, not a rule, and that it may pick a different page than you did. Redirects are the strongest signal, the canonical tag comes next, and inclusion in a sitemap is the weakest. When those signals disagree, for example the sitemap lists one address and the tag names another, Google has less reason to follow any of them, and its documentation says not to specify different canonicals through different methods.

The only place to see what Google actually decided is Search Console. Open URL Inspection for the page and read two fields: User-declared canonical, which is the tag this checker reads, and Google-selected canonical, the page Google chose as the authoritative version. If they differ, Google has overridden you, usually because the content of the two pages is not as similar as the tag claims, because internal links and the sitemap point at the other version, or because the named target redirects or errors. No outside tool, this one included, can read the second field; it lives in your account.

Syndicated articles and content libraries

This is where the tag decides the most for regulated firms. Many financial advisors, CPAs and insurance agents publish articles from a vendor's pre-approved content library, and the same article appears word for word on hundreds of other firm sites. For a set of identical pages, search engines will generally show one. The canonical tag on your copy is part of how that one gets chosen.

• Pointing at the vendor. If the platform that publishes the article inserts a canonical naming the vendor's own copy, that tag tells Google your page is a duplicate and the vendor's page is the original. Your copy is asking not to rank.
• Pointing at itself. Your copy claims to be the original, along with every other firm that received the same article. It may still be folded into someone else's version, but at least it is not volunteering to be.
• Missing. Google picks without any hint from you, and with hundreds of identical copies, your firm's is rarely the pick.

The practical step is short: if your articles come from a content library, check where the canonical points on two or three of them. If it names another domain, ask the vendor whether it can be changed, and treat the articles you want found as ones worth rewriting with your own experience, since an identical page has nothing to distinguish it whichever way the tag points. The same logic applies to a multi-office practice that publishes one disclosure or service page per location with only the city changed.

Canonical, noindex or redirect

Three tools solve overlapping problems, and mixing them up is common. A 301 redirect is for an address that should not exist any more: visitors and crawlers are sent to the new URL, and it is the strongest canonical signal there is. A canonical tag is for an address that must keep working for people, like a tracking link or a filtered listing, while search engines consolidate onto the clean version. A noindex tag removes a page from search results entirely. Google advises against using noindex to steer canonical selection within one site, because it blocks the page from Search completely rather than pointing its signals anywhere, and against using robots.txt for canonicalization at all. If a page carries both a noindex and a canonical to another URL, it is sending two different instructions; the Noindex Checker reads the other half.

Common mistakes

• The staging domain in production. A site built on staging. or a preview host ships with every canonical still naming it. The pages look fine and quietly ask Google to rank an address that is password-protected or gone. The target status row on this card catches this when the staging host answers with an error.
• A canonical that points at a 404 or a redirect. Pages move, templates keep the old address. A tag naming a missing page gives Google a hint it cannot follow.
• Every page canonicalised to the homepage. A theme setting or a hard-coded tag in a shared header. Every page on the site now claims to be a duplicate of the home page.
• http in the tag on an https site. The canonical names an address that redirects, which puts the redirect and the tag in disagreement about which URL is preferred.
• Two canonical tags. Usually a theme and an SEO plugin each printing one. If they disagree, Google may disregard both. This checker reads the first one only, so view the page source and search for canonical if you suspect a second.
• A canonical injected by JavaScript that differs from the served one. Google says a canonical added with JavaScript has to be injected properly; a served tag and a scripted tag that disagree are two conflicting signals.
• Paginated series all pointing at page one. Page two of a blog archive is not a duplicate of page one. Each page in a series should normally carry its own self-referencing tag.

How to add one

On WordPress, an SEO plugin almost certainly adds a self-referencing canonical already; the plugin's settings for a post let you override it when a page really is a copy of another. On a static or custom-built site, put the tag in the shared head template and build the href from the page's own final URL, with the scheme, the host you actually serve and the trailing-slash style your server redirects to. Then run this checker on a few representative pages: the home page, a post, a listing page and a URL with a tracking parameter added. The first three should read self-referencing; the last should point to its clean version.