Search
Tell search engines which address is the real one
Your page does not declare a preferred address, so the same content can be reached at more than one URL.
Why it matters
The same page is usually reachable several ways: with and without www, with and without a trailing slash, and with tracking parameters after a link is shared. Search engines see those as separate pages and split your reputation between them. A canonical tag says which one is the original.
How you would notice it
- Search results show your page under an unexpected address.
- Adding www to your address loads the same page but Google treats it as different.
- Pages linked from an email campaign rank differently from the same page found organically.
What to do
We make every page on our own site declare its address from one place. Two notes from setting that up.
Stage 1
- Decide the one address you want people to land on, and use it consistently. If your site serves www, use www everywhere.
- In the head section, add a canonical link pointing at that address.
The address here should be the final one a visitor lands on, not one that redirects somewhere else.
One declared origin, no drift possible
Every page on our site derives a single absolute address from one declared origin, combined with that page's own path at build time. If the origin is missing, the build stops instead of quietly shipping a page with no canonical. Because no page writes its own address, a bare-domain address and a www address can never both appear on the site, and a page cannot go live without this tag.
Source: Canonical origin derivation across our own built site
<head>
<link rel="canonical" href="https://www.example.com/services/">
</head>Stage 2
- Add the tag to every page, each pointing at its own address. A canonical that points every page at the homepage tells search engines your other pages are duplicates.
- Make sure your internal links use that same address, so you are not contradicting yourself.
Canonical and og:url are one claim
The canonical tag and the og:url tag state the same thing, so on our site both come from the same line of code, and we proved they cannot separate. In a test rebuild we removed the canonical, and the check comparing each page's og:url against its canonical failed in the same run, naming both. If you write these two by hand in different places, that is the failure you will eventually ship.
Source: Search signals negative test on our own rebuilt site
How to check it worked
Confirm it worked
Load your page and use the browser View Source. Search the page for the word canonical. You should find one link tag containing your preferred address. Re-run the Siege Test and confirm the Canonical URL check turns green.