The same page can be available at several addresses without it being immediately obvious. Sorting changes the order of products, a filter narrows the selection, and a campaign parameter records where a visit came from. Search engines may see these as separate URLs, even when their content is identical or very similar. A canonical URL identifies your preferred address for indexing, but it does not replace a coherent site structure.
1. Why similar content appears at different addresses
URL parameters are pieces of information usually added after the question mark in an address. An application can use them for filtering, sorting, campaign measurement or other functions. The technical SEO challenge arises when numerous combinations produce pages with no meaningful differences for search.
- Sorting options can display the same products in a different order, such as by price or name.
- Filters can create selections based on colour, size or availability. Some selections are genuinely useful and distinct; others differ very little.
- Campaign parameters, such as utm_source, identify the source of a visit without usually changing the main content.
- Technical variations, including HTTP and HTTPS or www and non-www addresses, can expose the same page if they are not handled consistently.
Duplicate content does not automatically mean a penalty. However, it can make it harder to select the right address for search results, split signals associated with the page and use crawling resources on variations that add little value.
2. What a canonical URL is and how rel canonical works
A canonical URL is the representative address for a group of identical or closely similar pages. A rel canonical declaration tells search engines which version you prefer. In HTML, the rel relationship with the value canonical and the href attribute are declared in a link element within the document head.
Visitors are not sent to another page. The address containing parameters can remain accessible, and sorting or campaign measurement can continue to work. The declaration concerns how search engines interpret the page, rather than how people navigate it.
It is useful for the preferred page to declare its own address as canonical, known as a self-referencing canonical. Prefer an absolute address with the correct protocol and domain to reduce ambiguity.
Canonical is a signal, not a binding instruction. Search engines assess it alongside redirects, internal links, sitemaps and content similarity. It does not guarantee that the destination will be indexed or that alternative URLs will immediately disappear from results.
3. Canonical or redirect: how to choose
The main difference is what happens to the visitor. A redirect sends the browser to another address. A canonical declaration keeps the page available at the requested address while expressing an indexing preference.
When a redirect makes sense
If an address has been permanently replaced and no longer serves a separate purpose, a permanent redirect, usually 301 or 308, is generally appropriate. Examples include moving a page and standardising HTTP/HTTPS or www/non-www variants where the configuration supports this.
When canonical makes sense
If a variation needs to remain available to users but its main content is equivalent, canonical may be suitable. A differently sorted listing can remain functional while identifying the original listing address as preferred.
Do not use canonical as a substitute for fixing incorrect addresses, errors or permanent moves. Redirecting away every parameter is not a safe universal rule either: it can interfere with filtering, campaign measurement or other application features.
4. Handling filters, sorting and campaign parameters
Start by separating parameters that only change presentation from those that substantially change the content. Applying one rule to every variation can hide useful pages or send inappropriate signals.
- Campaigns: if the parameters do not change the page, canonical can point to the address without the campaign tags. Check separately that visit measurement still works.
- Sorting: if only the order of the same set of items changes, the base address is often a suitable destination.
- Meaningful filters: a selection serving a distinct search intent may deserve its own indexable page, with a stable address and appropriate content.
- Combinations without distinct value: decide how to limit their generation and discovery. Canonical alone does not solve uncontrolled URL multiplication.
Take care with pagination. Successive pages in a listing may contain different products or articles. Do not automatically declare the first page as canonical for every subsequent page. Each paginated page should normally have its own canonical address.
If your goal is to exclude a page from the index, canonical is not the dedicated tool for that job. The noindex directive serves a different purpose. Avoid combining them without a clear intention: one requests exclusion, while the other proposes a representative address.
5. What to check in your site configuration
Work with a few representative examples: a standard page, a campaign-tagged version, a sorted version and a filtered version. These checks do not depend on any particular control panel.
- Inspect the canonical declaration. Check the HTML source and, if JavaScript changes the page, the rendered document. Look for conflicting declarations.
- Check the destination. Your preferred address should be accessible and indexable, with equivalent content rather than an error or an unrelated page.
- Avoid chains. Point directly to the final address, not a version that redirects or declares another canonical destination.
- Align your signals. Use preferred addresses consistently in internal links and include the canonical pages you want indexed in your sitemap.
- Check crawler access. If robots.txt blocks crawling of a variation, a search engine may be unable to read its canonical declaration.
- Review application rules. Check SEO settings, extensions, the theme, server rules and caching, particularly after changes or migrations.
On WordPress, themes and plugins can generate overlapping declarations. A WordPress hosting service provides the environment in which the site runs, but canonical rules still need checking in the actual site configuration. When migrating to another web hosting service, repeat the checks for protocol, domain and redirects.
6. Why a search engine may choose another canonical URL
A search engine may decide that another URL better represents the group of pages. This can happen when signals conflict, the declared destination is inaccessible or the pages are not sufficiently similar. Differences between the initial HTML and the rendered page can also complicate interpretation.
In search engine tools for site owners, compare the declared canonical with the search engine's selected canonical where that information is available. Inspect both pages and the links pointing to them. After you make corrections, the pages need to be crawled and processed again; the change is not instantaneous.
A practical takeaway
Choose a redirect when an old address needs replacing, and canonical when similar variations need to stay accessible. Group parameters by purpose, test representative examples and keep technical signals consistent. For questions about the hosting service, consult the frequently asked questions; indexing decisions need separate investigation at the site and search engine level.
