Cross-Domain Canonical and Content Syndication Without Duplicates

You publish an article on your own site. Then a partner, a news aggregator or a second site you own republishes it. Now the same text lives at two web addresses on two different domains, and search engines have to decide which one to show. A cross-domain canonical is a rel="canonical" link on one domain that points to a page on another domain. It looks like the obvious fix. For syndication, though, Google's advice is more specific, and getting it wrong can leave the copy in results while your original is the one that is filtered out.

This guide is for publishers, bloggers and site owners who either syndicate their content to other sites or run more than one domain with overlapping pages. You will learn how Google treats canonical signals, what it recommends for syndication partners, how to set up each option, and how to check which URL Google actually picked.

Nothing here can force a particular page to appear in search. Canonical tags and noindex are signals and instructions you give; search engines decide what to show.

Before you begin

Decide which case you are in, because the fix differs.

Case A: you own both domains and the pages are the same (for example an old domain and a new one, or a brand site and a regional site with identical pages).
Case B: a partner republishes your article on a domain you do not control.
Case C: someone copied your page without permission.

You need edit access to the page templates or HTTP headers of the pages you control. For Case B you also need a way to ask the partner for a change, ideally written into your syndication agreement. For checking results you need a Search Console property for your site.

Time: under an hour for the setup, plus waiting for recrawling. Google can take a while to revisit a page. The risk is low if you work on one section first and avoid combining signals that point in different directions.

How Google picks a canonical

When several URLs have the same or very similar content, Google groups them into a cluster and picks one as the canonical, the one it shows in results. Google's documentation lists the ways you can tell it which URL you prefer, from strongest to weakest:

Redirects. A strong signal. Google says to use a permanent redirect when you are deprecating a duplicate page.
rel="canonical" annotations. Also a strong signal. They can go in the page's <head> as a <link> element or in an HTTP response header, which suits non-HTML files such as PDFs.
Sitemap inclusion. A weak signal.

Combining signals increases the chance Google follows your preference, but none is required, and Google will choose a canonical on its own if you do not specify one. So a canonical is a strong hint, not a command. For more on same-domain canonicals, see our guide to canonical URLs and duplicate pages.

Google's documentation adds rules worth following whichever domain you use:

Use absolute URLs in the canonical, not relative paths.
Put a self-referencing canonical on the canonical page too.
Choose one method for a page (HTML or HTTP header) rather than both, because mismatches cause errors.
Do not use robots.txt for canonicalization, because a disallowed URL can still be indexed without content.
Do not use noindex to pick a canonical between versions on the same site, because it blocks the page from search entirely.

What Google says about syndication

Google's troubleshooting documentation for canonicals says rel="canonical" is not recommended for avoiding duplication with syndication partners, because the syndicated pages are often very different from the original: different layout, extra text, different surrounding content. The recommended approach is for the partner to block indexing of your content.

Google's help for news publishers gives the exact markup for partners. A Googlebot-News robots meta tag set to noindex keeps the copy out of Google News. A Googlebot robots meta tag set to noindex keeps it out of Google Search as well. Because the second tag affects Search, it is the one that addresses duplicate results in web search.

In plain terms, for Case B the instruction is: ask the partner to add noindex to their copy, not a canonical tag pointing at you. A partner may resist, because a noindex page cannot bring them search traffic. That is a business conversation, and your agreement is the place to settle it.

Step-by-step instructions

Step 1: Identify every copy

Search for a distinctive sentence from your article in quotation marks, and check Search Console for pages where Google selected a different canonical than the one you declared. Make a list of the copies and the domains they are on. Sort each into Case A, B or C.

What success looks like: You have a table of copies with the domain, the owner and the case.

Step 2 (Case A): Prefer a redirect when one domain is retiring

If both domains are yours and you no longer need the duplicate to be visible to visitors, a permanent redirect from the duplicate to the preferred page is the stronger option in Google's list. Use 301 redirects page by page, not a blanket redirect to the home page. If the migration involves a whole domain, follow the domain change guide.

What success looks like: Requesting the old URL returns a 301 response with a Location header pointing at the matching page on the preferred domain.

Step 3 (Case A): Use a cross-domain canonical when both pages must stay live

Sometimes both domains need to remain accessible, for example a store that must keep a legacy domain for existing customers. Put a canonical on the duplicate that points to the preferred page on the other domain:

<link rel="canonical" href="https://www.example.com/guides/anchor-text/">

Or, for a PDF, send it as an HTTP header:

Link: <https://www.example.com/downloads/guide.pdf>; rel="canonical"

The preferred page should carry a self-referencing canonical to its own address. Keep the two pages the same in substance. A canonical is meant for duplicates, not for pointing one page at a different one.

What success looks like: The duplicate page source shows one canonical to the other domain, and the preferred page has a self-reference.

Step 4 (Case B): Put `noindex` in the syndication agreement

Ask each partner to add this to the head of the republished article:

<meta name="Googlebot" content="noindex">

If the partner only needs to stay out of Google News, Googlebot-News is the narrower tag. State clearly which one you need. Also ask for a visible credit and a link to your original article. Google's documents on this subject do not require that link, so treat it as a courtesy and a path for readers, not as a ranking tactic.

Make sure the partner does not block the page in robots.txt. For noindex to work, Googlebot must be able to crawl the page and see it. Google states that a page blocked by robots.txt never has its noindex read, and it can still appear in results if other pages link to it.

What success looks like: Partner pages contain the noindex meta tag in the HTML Google receives, and the pages are not disallowed in robots.txt.

Step 5 (Case C): Handle copies you did not authorize

Syndication is a deal; a copy made without permission is a different problem. Google's troubleshooting guidance suggests contacting the host of the copying site, or filing a DMCA request with Google. Do not try to solve it by adding signals to your own page, because you cannot make another site obey them.

What success looks like: You have contacted the host or filed the request and kept a record.

Step 6: Check which canonical Google selected

In Search Console, use the URL Inspection tool on your original and on the copies you control. In the indexed results, the page indexing section shows the user-declared canonical and the Google-selected canonical. They should match for pages you control. Google notes that a live test can show only your declared canonical and cannot predict Google's choice, so use the indexed view for the answer. The value can also lag by a few hours.

If the selected canonical is not what you expected, check for conflicts: a redirect that points elsewhere, a canonical set by a plugin, or a different canonical in the HTTP header than in the HTML. Google says it may keep pages in a duplicate cluster for up to two weeks after you fix content problems, so allow time.

What success looks like: For pages you control, the declared and Google-selected canonical match, or you understand why they differ.

When the wrong domain wins

Sometimes Google chooses the copy as canonical. Possible causes are a partner page that differs from your original, or a mismatch in your own signals, such as a redirect or plugin that sets a different canonical. Google also warns about two unusual causes. Hosting misconfigurations can serve one domain's content at another domain's URL, and malicious code can inject a cross-domain canonical or redirect to spam. If your canonical points somewhere you did not put it, treat it as a possible compromise and inspect your HTML source and server headers.

Another limit: if Google selects a canonical in a Search Console property you do not own, you cannot see performance data for the duplicate page in your own account.

Final verification checklist

Each copy of the content is listed with its domain, owner and case.
Retired duplicates on domains you own redirect with 301 to the matching page.
Duplicates you keep live carry one absolute canonical URL to the preferred page.
The preferred page has a self-referencing canonical.
No page uses both an HTML canonical and a different HTTP header canonical.
Partner copies carry a noindex robots meta tag and are not blocked in robots.txt.
URL Inspection shows the Google-selected canonical you expect for pages you control.
Your sitemap lists only the preferred URLs.

Common mistakes to avoid

Asking syndication partners to canonical to you. Google does not recommend this, because syndicated pages often differ. Ask for noindex instead.

Blocking the syndicated page in robots.txt. Googlebot then cannot read the noindex. Allow crawling and use noindex.

Canonicalizing to a page with different content. A canonical is for duplicates. Point to the page that actually matches.

Using relative canonical URLs. Google asks for absolute URLs. Include the scheme and host.

Setting an HTML canonical and a different header canonical. Pick one method, and make sure it is consistent.

Using noindex on the original to choose a canonical. That removes the page from search. Use redirects or canonicals on duplicates, not noindex on the page you want shown.

Expecting the change to apply immediately. Google has to recrawl. Use URL Inspection for key pages, and avoid repeated requests.

Frequently asked questions

Does Google penalize syndicated content?

Google's documentation on canonicals describes duplicate content as something it groups and picks one version of. The question is which copy is shown, not a penalty. This guide cannot tell you how any single page will rank.

Can I use a cross-domain canonical for syndication anyway?

You can, but Google's guidance says it is not recommended for syndication partners, and that blocking indexing on the partner side is the most effective method. A canonical may still be taken as a signal, but you should not rely on it.

Should the partner link back to my original?

A link back is a sensible request, and it helps readers find the source. Google's pages we reviewed do not make it a requirement, so do not treat it as a sure fix for duplicate results.

What if the partner refuses to use noindex?

Then you cannot control which copy appears. Consider changing what you syndicate, for example an excerpt with a link instead of the full article, or choose partners who accept your terms.

How do I see which URL Google chose?

Use URL Inspection on the page and read the Google-selected canonical in the indexed results. The live test shows only the canonical you declared.

Official references

Next step: See how Crawvix checks a website and capture a baseline once scanning opens. This guide provides technical information, not a guarantee of rankings, traffic or index inclusion.