Is a canonical the same as a redirect?

No. A redirect sends the user and the crawler somewhere else. A canonical leaves both on the page and expresses a preference about which URL should be credited. Use a redirect when the URL should not be reachable, and a canonical when it should be reachable but should not be the one that ranks.

Can I canonicalise across domains?

Yes, and it is the correct tool for syndicated content: the copy points at the original. It relies entirely on the engine agreeing with you, so it works best when the original is clearly the earlier and stronger version.

Does a canonical pass link equity?

It consolidates signals onto the target, which is the practical effect people mean. It is not a guarantee and it is not a redirect. If the equity genuinely matters and the URL has no reason to stay reachable, redirect instead.

Canonical tags: what they do, and the four ways they backfire

You have canonical tags and want to be sure they are consolidating duplicates rather than quietly hiding pages.

Short answer

A canonical tag tells a search engine which URL you consider the master copy of near-duplicate content, so signals consolidate on one URL instead of splitting. It is a hint, not a directive: engines can and do ignore it when the evidence disagrees. Pointing a unique page at a different URL is the fastest way to remove that page from search without meaning to.

What problem it solves

The same content reachable at several URLs splits its links and its history across all of them, so none ranks as well as one would. A canonical nominates the version that should accumulate everything. That is the whole job, and it only applies where the pages really are duplicates or very close.

Mistake one: canonicalising unique pages to the homepage

This happens when a template sets the canonical from a variable that is empty on some routes. Every affected page tells the engine it is a copy of the homepage, and the engine, taking a reasonable reading, drops them. It is invisible in a browser and it can remove an entire section.

Mistake two: relative or wrong-protocol URLs

A canonical should be absolute and should match the URL you actually serve, including protocol and host. A canonical pointing at the http version, or the non-www version, when you serve https and www, tells the engine your live URL is not the one you want, which at best wastes a redirect and at worst splits everything again.

Mistake three: chains and loops

Page A canonicalises to B, B canonicalises to C. Engines may follow one hop and give up on more. A canonical should point at a URL that is itself canonical, returns 200, and is indexable. Auditing this across a site is tedious by hand and trivial with a crawl.

Mistake four: contradicting yourself

A canonical pointing one way while your internal links, sitemap, and redirects all point the other is a conflicting set of signals, and the engine resolves it however it likes. Consistency across those four is what makes a canonical get honoured; the tag alone is the weakest of the four.

Self-referencing canonicals

Harmless, mildly useful, and the sane default. A page pointing at itself removes ambiguity when the URL picks up tracking parameters, and it means a template bug that empties the variable is more likely to be caught than to quietly point everything at the homepage.

How to check

Fetch the page as a crawler would and read the canonical from the raw HTML. If it is injected by JavaScript, a crawler reading the first pass will not see it, which for canonical purposes is usually the same as not having one.

Questions

Is a canonical the same as a redirect?: No. A redirect sends the user and the crawler somewhere else. A canonical leaves both on the page and expresses a preference about which URL should be credited. Use a redirect when the URL should not be reachable, and a canonical when it should be reachable but should not be the one that ranks.
Can I canonicalise across domains?: Yes, and it is the correct tool for syndicated content: the copy points at the original. It relies entirely on the engine agreeing with you, so it works best when the original is clearly the earlier and stronger version.
Does a canonical pass link equity?: It consolidates signals onto the target, which is the practical effect people mean. It is not a guarantee and it is not a redirect. If the equity genuinely matters and the URL has no reason to stay reachable, redirect instead.

Measured, not asserted

On ihatepdf.cv, the non-www and www versions of the homepage both accumulated clicks, appearing as separate rows in the same Search Console export. Splitting one page across two hosts is the ordinary version of this mistake, and it happened on a site that reached 100,542 users.

Free tool for this: Crawler View. No account, nothing uploaded.

Where this goes deeper

Every number on this page comes from one complete dataset: one product taken from zero to 100K+ users on search alone, with nothing spent on advertising. The full argument is Chapter 10 of the book. Five chapters are free to read.