Canonical tag
A hint about which address is the real one, used when copies of a page are hard to avoid.
What does Canonical tag mean?
A hint in a page's code that names the preferred address when the same content can be reached at more than one URL. Visitors may still be on the other address.
Also called: rel canonical, canonical URL
A preference, not a move
If the same page opens with and without www, or with and without a trailing slash, you have two addresses. A canonical tag on each copy can point at the one you want indexed. The visitor is not automatically sent there. A redirect is the tool that actually moves them.
Search engines treat the tag as a hint. They can ignore it if the pages are not really duplicates. Use it for true copies, not as a way to funnel unrelated town pages onto one URL.
Where duplicates come from
Printer views, tracking parameters on ad links, and a builder that serves both a http and an https version are typical sources. The clean fix is often one public URL, with the extras redirected. The tag covers the leftovers you cannot redirect, such as a necessary filter.
A syndicated article that you also host can point its canonical at the original, or the original can point at you, depending on who should be found. Do not publish two competing originals and hope the tag settles an argument you have not had.
What you should ask
Ask which URL is canonical for the homepage and for each service, and whether the other versions redirect. You should not have to read the HTML. If a plugin offers a canonical field on every screen, leave it blank unless you know of a duplicate.
We set a single preferred host, HTTPS, and a consistent path, then redirect the variants. The tag is there for the cases redirect cannot cover. Indexing is what you are trying to keep unambiguous.
An example
A gym's class page was linked in an email with a long tracking tail, and that tail started showing up as if it were a second page. The canonical pointed back at the clean class URL. The email tracking still worked. Search did not grow a duplicate.
Questions we get asked
Should every page point its canonical at the homepage?
No. That tells crawlers the homepage is the only real page. Each page should name itself, unless it is a genuine duplicate of another URL.
Can I see the tag?
It is in the page code, not on the screen. You do not need to view source to run the business. Ask us if a URL is multiplying. Search Console's page report is the symptom list.
