How a self-referencing canonical works
A canonical tag is a line in a page’s head naming the address the page should be indexed under. The self-referencing case is the simple one: the page names itself. The address should be written in full, and it should be the exact version you want indexed — the right protocol, with or without www as you have chosen, the correct casing, and the trailing slash if your site uses them.
The tag is a strong hint rather than an order. Google weighs it against internal links, sitemaps, redirects and the content itself when deciding which of several similar addresses is the real one. A page that names itself is casting a clear vote in that decision. A page that says nothing leaves the choice entirely to the search engine, which may pick a version you did not want.
Why self-referencing canonicals matter
Almost every site serves the same page under more addresses than its owner realises. Campaign tracking parameters, click identifiers that ad platforms append on arrival, filters and sort orders, session identifiers, print versions and stray capital letters all produce distinct addresses carrying identical content. Each one is a candidate for indexing, and the evidence that should support one page ends up spread across several.
The tag also travels with your content. When another site copies a page and keeps the markup intact, the canonical still points home, which helps search engines work out where the original lives. That is not a guarantee, but it costs nothing and it is one of the few defences you have against casual copying.
Where self-referencing canonicals go wrong
The damaging configuration is a template that points every page at the homepage. It reads as a declaration that the whole site is a duplicate of the front page, and it can quietly strip the rest of the site out of the index. Anyone inheriting a site should check for this before anything else.
The quieter faults are mismatches. A canonical naming an address that now redirects, or one that no longer exists, sends a confused signal. So does a page that names itself while also carrying a noindex directive, since the two instructions contradict each other. Two plugins both adding a canonical is common as well, and conflicting tags tend to be ignored altogether — which leaves you back where you started.
How to act on it
First find out what you already have. WordPress outputs a self-referencing canonical on single pages and posts by default, and most SEO plugins add one too, so the job is usually verification rather than installation. Use the URL Inspection tool in Google Search Console and compare the canonical you declared with the one Google actually selected; a difference is worth understanding before you change anything.
Then test the variants. Add a tracking parameter to a live page and confirm the canonical still names the clean address, and check a filtered or sorted view of a listing the same way. Fixing these is straightforward technical SEO work, and it is usually the cheapest way to reduce duplicate content on an established site.
