Web design and digital-business ideas grounded in real tasks.

A canonical link is a preferred address, not a redirect

A web designer viewing two similar generic product pages on monitors.
AI-generated photographic illustration; not documentary photography.

You open a product page through a tracking link. The address bar keeps the extra parameter, yet the page source names the clean product address as canonical. Nothing has gone wrong simply because the browser did not move. A canonical link and a redirect do different jobs.

A canonical annotation expresses a preference about which address represents duplicate or very similar content. It does not, by itself, tell the browser to navigate elsewhere.

Follow one visitor through two cases

In an illustrative shop, /mug/?campaign=autumn displays the same mug as /mug/. The parameter page can remain available while its canonical link points to the clean address. The visitor still sees the parameter in the address bar.

Now imagine an old address, /old-mug/, that the shop has permanently replaced. A server redirect can send a visitor to /mug/. The browser follows the response and changes its address. That is observable navigation, not merely a preference written in the document.

Check the result, not just the setting

Google’s canonicalization documentation treats redirects and canonical link annotations as strong signals. Signals do not amount to a guarantee that a preferred URL will be selected. Consistent internal links and sitemap entries help communicate the same preference.

For a small site check, open the original address and record where the browser ends up. Then inspect the final page’s canonical annotation. These are two separate observations. If a redirect chain exists, record each step rather than assuming the first and final addresses tell the whole story.

A common configuration mistake is to nominate one URL in the page source but a different version in navigation or the sitemap. Another is to point unrelated articles at one canonical target merely because they share a topic. Canonicalization concerns duplicate or closely equivalent content, not a general-purpose way to rank one page above another.

Before changing anything, decide the desired reader experience: should both addresses keep serving the page, or should visitors move to one replacement? That question makes the technical choice much clearer.