Canonical Tag Builder

It is a hint about which of several URLs should be indexed, not a redirect. Visitors can still reach any of the others, and the tag does not stop them. That distinction matters more than it sounds: many people add a canonical tag expecting it to fix a redirect chain, when in fact a tag pointing at a redirecting URL is ignored, and the fix is to remove the extra hop so the target returns 200 directly.

It is also per page. One page gets one canonical tag, pointing at the URL you want indexed. Cross-page canonicals — where /page-a points at /page-b — are for genuine duplicates such as a print view or a paginated listing, and they consolidate ranking signals rather than redirecting anyone.

Most canonical problems are really disagreement about one of four things. Scheme: the same page on http and https is two URLs, and the tag should name the https one with the other redirected, not tagged. Host: example.com and www.example.com are separate URLs unless one redirects to the other, and the tag settles which is the real one. Trailing slash: /page and /page/ are different paths to most servers, which is why the tag needs to match whatever your routes actually do. Query string: parameters that select content belong in the canonical URL, and parameters that identify a campaign click do not.

Each of those is a dropdown here, so the tag reflects your decision rather than a guess. The output then lists the resulting variants so you can confirm the others really are the same page.

Four situations cause a canonical tag to have no effect, and all four are worth checking before you assume a bug. The target URL redirects, so the crawler never sees a tag on a 200 response. The target is blocked in robots.txt or marked noindex, so it cannot be selected as canonical. The tag is relative rather than absolute and the crawler cannot resolve it, or it is malformed, commonly a missing closing slash. Or the page has two canonical tags pointing at different URLs, which leaves the crawler to choose and produces exactly the inconsistency the tag was meant to remove.

There is also the hreflang interaction: if you have hreflang annotations, each needs its own self-referencing canonical, and they must agree with each other.

Ad platforms and analytics append parameters like utm_source and gclid to links they send, which means the same page gets indexed under dozens of URLs that differ only in a campaign identifier. Each is a distinct URL to a crawler, and together they dilute the signals for the page you actually want ranked. The tool lists the campaign parameters it finds so you can strip them from the canonical URL.

It reports them rather than removing them silently, because a query string is sometimes genuinely part of the page — a filtered product listing or a paginated view can depend on its parameters, and stripping those from the canonical tag would merge pages that are not the same.

Copy the tag into the <head> of the page it describes, not onto every page. It should be self-referencing: the canonical on /page points at /page. Use one per page, and if the page already has a tag, replace it rather than adding a second.

After deploying, confirm the target returns 200 and is indexable, and that the variants listed here resolve to the same content. If one of them returns something different, a canonical tag cannot reconcile that, and the real fix is at the routing or redirect layer.