Canonical Tag and Duplicate Pages: How to Find and Fix Duplicates on Your Website
Duplicate pages are one of the most common technical issues I find in audits: the same page loads at several addresses, and Google decides on its own which one to show. Below - where duplicate pages come from, when a canonical tag is enough and when you need a 301 redirect or noindex, how to find duplicates with Search Console and Screaming Frog, and which canonical mistakes to avoid.
Contents
- What duplicate pages are and why they hurt
- Where duplicate pages come from: common sources
- Your tools: canonical tag, 301 redirect, noindex and hreflang
- Table: which situation needs which tool
- How to find duplicate pages: 4 steps
- Common canonical tag mistakes
- Checklist: duplicate pages under control
- Frequently asked questions
- Summary
What duplicate pages are and why they hurt
Duplicate pages are different URLs with the same or nearly the same content. To a visitor it is one page. To Google it is several separate addresses, and it has to pick one canonical URL - the main version for the index and search results.
Duplicate content on its own does not trigger a penalty: Google states clearly that it does not penalize technical duplicates unless they are an attempt to manipulate results. The real problems are different:
- Wasted crawl budget. Googlebot crawls dozens of variants of one page instead of new products and articles. For a site with a few hundred pages this is rarely critical, for an online store with filters it is noticeable;
- Split signals. Links, internal linking and user signals are shared across several URLs instead of working for one page;
- The wrong page ranks. Google may pick a URL with a UTM tag, a print version or a sorted listing as canonical instead of the main category;
- Pages compete with each other. Two similar pages targeting one query get in each other's way, and rankings fluctuate.
The canonical tag is not a command. Google weighs it together with redirects, internal links, sitemaps and HTTPS, and may choose a different canonical page if those signals contradict each other.
Where duplicate pages come from: common sources
Most duplicates are created not by content editors but by the CMS, the server or marketing tools. Check these first:
- www and non-www. example.com and www.example.com both return the same page with a 200 status;
- http and https. The old http version is still reachable without a redirect to https;
- Trailing slash. /catalog/ and /catalog serve identical content;
- Parameters and UTM tags. ?utm_source=..., ?ref=..., session IDs, ad tracking parameters;
- Filters and sorting in e-commerce. ?sort=price, ?color=black&size=10 - thousands of combinations with almost the same set of products;
- Pagination. /catalog/?page=1 duplicates /catalog/;
- Print and AMP versions. /print/ or ?print=1 with the same text;
- Products in several categories. /men/sneakers/model-x/ and /sale/model-x/ - one product page at two addresses;
- Language and regional versions. Pages for the US, UK and Australia with the same English text;
- Letter case and index files. /Catalog/ and /catalog/, /index.php and / as separate addresses.
Your tools: canonical tag, 301 redirect, noindex and hreflang
There is no one-size-fits-all fix. The choice depends on whether people actually need the duplicate page.
rel=canonical
A tag in the page <head> that points to the preferred canonical URL: <link rel="canonical" href="https://example.com/catalog/">. Both pages stay available to visitors, while Google consolidates signals on the canonical one. For PDFs and other non-HTML files you can send the canonical in a Link HTTP header. Use absolute URLs and one canonical per page.
301 redirect
The strongest signal: the duplicate simply stops existing, and both users and bots land on the main URL. Use it for www/non-www, http/https, trailing slashes, letter case and old URLs after a migration. If visitors do not need the duplicate page, redirect it rather than adding a canonical.
noindex
The <meta name="robots" content="noindex"> tag removes a page from the index but does not pass signals to another page. It is for pages that should not appear in search at all: internal search results, the cart, utility filters. Google advises against using noindex to choose a canonical version and against combining noindex with a canonical tag on the same page.
Parameter handling
Search Console no longer has a URL Parameters tool - Google retired it in 2022. Parameters are now handled on the site itself: a canonical from parameterized URLs to the clean address, internal links without UTM tags, and only canonical URLs in the sitemap. Blocking parameters in robots.txt saves crawl budget, but Google then cannot see the canonical on those pages, so use it selectively.
hreflang for language versions
Language versions are not duplicates, even if the texts are similar. Each one gets a self-referencing canonical, and hreflang links the versions together. More on this in the article on international SEO.
Table: which situation needs which tool
| Situation | What to use |
|---|---|
| www and non-www, http and https | 301 redirect to the main version |
| Trailing slash, letter case, /index.php | 301 redirect to one URL format |
| UTM tags, ref, session IDs | Canonical tag to the clean URL, internal links without parameters |
| Sorting and filters with no search demand | Canonical tag to the category or noindex |
| Filters with demand ("black running shoes size 10") | A dedicated landing page with its own title, copy and self-referencing canonical |
| Pagination | Self-referencing canonical on every page, ?page=1 - 301 to the category |
| Print version | Canonical tag to the main page |
| Product in several categories | One primary product URL, canonical to it from the other addresses |
| Language and regional versions | Self-referencing canonical + hreflang |
| Cart, internal search, account pages | noindex |
A separate note on online stores: filters are the biggest source of duplicates and, at the same time, the biggest growth opportunity. How to split filters into landing pages and technical pages is covered in the article on e-commerce SEO.
I will find the duplicates on your site
Send me your website address - I will check the indexing report, canonicals and redirects and tell you which duplicates hold back your rankings and how to fix them.
Message me on TelegramHow to find duplicate pages: 4 steps
Page indexing report in Search Console
Open Indexing - Pages and review the reasons why pages are not indexed. Three statuses matter for duplicates: "Duplicate without user-selected canonical" - no canonical is set and Google chose one itself; "Duplicate, Google chose different canonical than user" - your canonical was ignored; "Alternate page with proper canonical tag" - everything is fine, this is expected behavior.
URL Inspection in Search Console
For a specific page, the URL Inspection tool shows two fields: User-declared canonical and Google-selected canonical. If they differ, look for the signals that contradict your canonical: internal links, the sitemap, redirects.
Crawl with Screaming Frog
Crawl the site and check the Canonicals tab (missing, non-indexable canonical, canonicalized to another URL), plus Content - Exact Duplicates and Near Duplicates (near-duplicate detection has to be enabled in the configuration). Also filter pages with identical titles and H1s - a quick way to spot duplicates the CMS "forgot" about.
site: search in Google
Queries like site:example.com inurl:utm, site:example.com inurl:sort or site: with a fragment of unique text in quotes show which URL variants made it into the index. The site: operator gives approximate data, so use it for a quick check, not for counting.
Finding duplicates is a standard part of a technical site review. What else such a review covers is described in the article on the SEO audit.
Common canonical tag mistakes
- Canonical pointing to a noindex page. A contradictory signal: "this is the main version, but do not index it". As a result, Google may ignore both instructions;
- Canonical chains. Page A points to B, and B points to C. Every canonical should point straight to the final canonical URL;
- All paginated pages canonicalized to page 1. Page 2 and beyond list different products, so they are not duplicates. This setup can leave products on deeper pages poorly indexed;
- Canonical pointing to a redirect or a 404. Only point to URLs that return 200 and are canonical themselves;
- Every page canonicalized to the homepage. A common result of a template error: every page on the site "asks" Google to index only the homepage;
- Relative URLs and http in the canonical. On an https site, a canonical with http or without the domain causes confusion;
- Two canonicals on one page. One added by the CMS, another by an SEO plugin. Google may then ignore both;
- Canonical contradicts other signals. The sitemap and internal links use one URL, the canonical another. The signals have to match.
Duplicates often show up in meta tags too: identical titles on different URLs are the first sign of a problem. How to write a unique title and meta description for every page is covered in the article on title tags and meta descriptions.
Checklist: duplicate pages under control
- The site resolves to one address: 301 from http to https and from the secondary www/non-www version
- One URL format: with or without a trailing slash, lowercase, no /index.php
- Every indexable page has a single canonical tag with an absolute https URL
- The canonical points to a page that returns 200, with no noindex and no redirect
- Paginated pages have self-referencing canonicals, not canonicals to page 1
- Filters are split into landing pages (self-referencing canonical) and technical ones (canonical to the category or noindex)
- Internal links and the sitemap contain only canonical URLs without UTM tags
- Language versions have self-referencing canonicals and correct hreflang
- Duplicate statuses in the Page indexing report are reviewed once a month
Frequently asked questions
What is a canonical tag in simple terms?
It is a tag in the page code that tells Google which URL to treat as the main one among several identical or similar pages. Google usually follows it, but may choose another page if other signals on the site contradict the canonical.
Does Google penalize duplicate pages?
No, there is no penalty for technical duplicates. But duplicate pages waste crawl budget, split links across several URLs and can put the wrong page in search results, so they are worth fixing.
Which is better: a canonical tag or a 301 redirect?
If visitors do not need the duplicate page, use a 301 redirect - it is the stronger signal. If the page has to stay available (UTM tags, sorting, print version, a product in several categories), use a canonical tag.
Do I need a canonical tag if a page has no duplicates?
Yes, a self-referencing canonical is good practice. It protects against duplicates that appear later: when links with UTM tags or parameters point to the page, Google immediately sees which URL is the main one.
Why does Google ignore my canonical tag?
The most common reasons: the pages are not actually duplicates, internal links point to the non-canonical URL, the sitemap lists a different URL, or the canonical points to a page with a redirect or noindex. Check the page in URL Inspection and align all signals on one address.
Summary
Duplicate pages do not lead to penalties, but they waste crawl budget and split signals across several addresses. Remove technical duplicates (www, trailing slash, letter case) with a 301 redirect. Mark pages that people need but search does not with a canonical tag. Close utility pages with noindex and connect language versions with hreflang. Then check the result in the Page indexing report: that is where you see whether Google agreed with your canonical.
Let me clean up your site's indexing
A technical audit of duplicates, canonical, redirect and hreflang setup, and indexing monitoring in Search Console. The first step is a free express audit, monthly SEO starts from $450/month.
Send a request