3 min read
The canonical tag on a page changed - does that matter?
By CrawlWarden, which builds the monitoring described at the foot of this article.
Published , sources last re-read .
Yes, it matters. A <link rel="canonical"> tag tells Google which URL is the real version of a page. Point it at a different URL and Google will usually consolidate the page’s ranking signals into that other URL and stop showing this one. Nothing on the page looks wrong. The three failure modes are a canonical pointing at the wrong page, a canonical pointing at a URL that no longer returns 200, and a page that silently stops pointing its canonical at itself.
The three ways canonicals break
| Failure | What Google does | Usual cause |
|---|---|---|
| Points at an unrelated URL | Consolidates this page into that one. This page stops appearing. | A template variable resolving to a default, or a copy-pasted head block. |
| Points at a URL that 404s or redirects | Treats it as a strong hint it cannot honour, and falls back to its own guess. Behaviour becomes unpredictable. | The target page was deleted or its slug changed, and the canonical was not updated. |
| Stops being self-referential | Depends what it now points at - but a page that used to declare itself canonical and no longer does has had something change. | A migration, a trailing-slash change, an http/https or www switch applied in one place and not the other. |
A canonical is a hint, not a directive
Google’s own documentation is explicit about this: the canonical tag is one of several signals it uses to choose a canonical URL, and it can disagree with you. Search Console reports its decision under "Google-selected canonical" alongside "User-declared canonical" in URL Inspection.
The practical read: a wrong canonical is not guaranteed to hurt, but a canonical that disagrees with the rest of your signals is a coin toss you did not need to take.
Pick one URL form and enforce it everywhere
Trailing slash or no trailing slash; www or bare; http or https. Each pair is two URLs as far as a crawler is concerned, and if both are reachable and both are indexable, authority splits between them.
- Redirect one form to the other at the server, with a 301. Server-level rules fire before the application boots and keep working when it does not.
- Make the canonical tag on every page point at the surviving form, absolutely, including the scheme and host.
- Make the XML sitemap list the same form. A sitemap that lists the redirected form is a slow, self-inflicted crawl tax.
The checks that report this
CrawlWarden watches a list of URLs you choose and reports what changed. It does not diagnose and it does not score. Severity is a property of the check rather than a setting: the number below is how many consecutive runs must see the new value before it becomes an email, and it is never less than two, because one observation is noise.
Canonicalconfirmed over 3 consecutive runs
Reports when the canonical URL on a page changes to a different target.
Canonical resolvesconfirmed over 3 consecutive runs
Fetches the canonical target and reports when it stops returning 200.
Canonical self-referentialconfirmed over 3 consecutive runs
Reports when a page that used to point its canonical at itself stops doing so.
Every check is available on every plan, including the free one. CrawlWarden pricing, per site rather than per URL.
Questions
- Should every page have a self-referential canonical?
- For a normal content page, yes - it is the cheapest way to remove ambiguity, and it costs nothing. The exceptions are deliberate: paginated series, syndicated copies and parameterised variants that genuinely belong to another URL.
- Does a wrong canonical always cost me the ranking?
- Not always. Google treats the tag as a signal, not a command, and will sometimes override it. But you are relying on Google noticing your mistake, which is not a plan.
- How fast does a canonical change take effect?
- On the next crawl and re-processing, so days to weeks depending on how often the URL is visited. Reverting it is the same wait in the other direction.
- Why does CrawlWarden confirm this over three runs rather than two?
- Canonical values are the most common thing a CDN serves inconsistently between cache nodes. Three consecutive observations of the new value separates a real edit from a stale variant.
Related
- redirect chains, which break canonical targets the same way
- the clicks drop a canonical edit produces days later
- sitemap membership, which should list the same URL form
- a page going noindex
- rich results vanishing after the same kind of template edit
- every failure mode CrawlWarden watches for
- every check it runs, and what a change in it means
Sources
What was checked, and where. An entry with no link is something that could not be verified, named here rather than left out.
Google Search Central: What is URL canonicalization
The sentence the third section rests on: "Google may choose a different page as canonical than you do, for various reasons. That is, indicating a canonical preference is a hint, not a rule."
Google Search Central: How to specify a canonical URL
The methods that express a canonical, and that Google consolidates ranking signals onto the URL it picks.
Not verified: how often Google overrides a declared canonical
Google says it may and gives reasons; it publishes no rate, and no reliable third-party measurement was found.
Not verified: search volume for this query
No keyword volume tool is connected to this project. Demand was judged from a live depersonalised US SERP.
Know the moment this happens to you
Give CrawlWarden a domain, keep the pages it suggests, and the first check runs immediately. Free forever for one site.