In short
- The canonical tag tells Google which URL to consider as the reference among several similar pages.
- It is a strong hint, not a directive: Google can choose another URL if the signals contradict each other.
- A facet you want to rank must have a canonical pointing to itself, never to the parent category.
- Paginated pages each carry their own canonical, not the one of page 1.
On an e-commerce site, the same page is often accessible under several addresses: with a sorting parameter, a tracking code, a color variant, a different category path. The canonical tag tells Google which one is the reference. Used well, it consolidates signals. Used badly, it makes pages disappear from the index.
What the canonical does
Placed in the head of a page, the <link rel="canonical" href="..."> tag indicates the URL you consider the reference version. Google uses it to group similar URLs and concentrate signals (links, relevance) on a single one.
A key point: it is a strong hint, not a directive. Google combines it with other signals, such as redirects, internal links and the sitemap. If these signals contradict each other, it may choose a URL other than the one you declared.
Good uses in e-commerce
| Situation | Recommended canonical |
|---|---|
| Category with sorting or tracking parameters | To the clean category URL |
| Facet you want to rank | To itself |
| Facet combination with no value | To the parent category, or crawling blocked |
| Paginated pages (page 2, 3...) | Each to itself |
| Product accessible under several category paths | To the main product URL |
| Variants of a product on distinct URLs | Depending on your strategy: one reference URL, or each variant if it answers its own search |
The mistakes that cost
The facet canonicalized to its category
This is the most frequent mistake: a filter module that automatically adds a canonical to the parent category on all facets, including those you want to rank. As a result, "160x200 linen fitted sheet" can never rank, since you are asking Google to ignore it in favor of the general category. Check the canonical of every strategic facet.
Page 1 as the canonical for the entire pagination
Google explicitly advises against it: every paginated page must carry its own canonical. See our guide to pagination.
Contradictory signals
A canonical to URL A, but internal links and a sitemap pointing to URL B: Google may keep B. Align the canonical, the internal links and the sitemap on the same URL.
The canonical used instead of a real decision
Two categories competing for the same query are not always solved with a canonical. If both pages are useful, they need to be differentiated; if one is superfluous, redirect it. See our guide to cannibalization.
Checking your canonicals
A crawler lets you list the canonical of each page and spot strategic categories and facets that point somewhere other than themselves. In Search Console, the URL inspection tool shows the canonical Google kept: if it differs from yours, look for the contradictory signal.
Once the right pages are declared as references, they still need tags and copy that live up to it: that is the work unblind automates for every indexed category and facet.
Frequently asked questions
Is the canonical tag mandatory?
No, but it is recommended on an e-commerce site, where the same page is often accessible under several URLs (sorting parameters, tracking, variants). A self-referencing canonical on every indexable page avoids ambiguity.
Does Google always respect the canonical tag?
No. Google treats it as a strong signal among others (redirects, internal links, sitemap). If these signals contradict each other, for example if your internal links all point to another URL, Google may keep a different URL.