Spammy Google Selected Canonical In Search Console

You can use the URL inspection tool on GSC (Google Search Console) to check the crawl/index status of a URL. The report contains a "Google-selected canonical" value that points to the actual webpage that Google chose for the content. Google can & will display some other webpage from the same or an external website in some cases. Sometimes the URL can point to a spammy or dangerous site.

Dan Lauer wrote on BlueSky, "I have never seen Google selected canonical to be a spammy site before like this? This is for a new clients blog and its not indexed yet. I discovered this inspecting it in GSC."

Here is the screenshot from the post:


John Mueller from Google responded:

These things are impossible to say without more details. Sometimes domains have used the same interstitial / parked page, which can easily lead to this. I'd check back after a few weeks to see how it settles down, otherwise post in the help community with the full details.

Click to expand...

I have personally faced such a situation multiple times and here is what I noticed:

1. If you have multiple home pages for a website with each pointing to a different language using hreflang attributes, Google may treat all these pages as duplicates of the main home page.
2. If an authority site or a parasite copy-pastes the exact same contents of your webpage, Google may treat that as the selected canonical to your webpage and choose to display that webpage in SERPs instead of yours.
3. If a document or media from your webpage is uploaded to an authority site (for example, pdf to archive.org or video to youtube), Google will choose the authority site.

If it happens to your webpages, you'll receive “Duplicate, Google chose different canonical than user” message in GSC. You can check the URLs and try to fix them by adding a canonical (if it doesn't exist) or by updating the content.

A quick way to tell them apart is to inspect a few other URLs on the site. If they all point to the same external canonical URL, then it's leftover clustering from the parked state. If only one page's content was duplicated, then yes, it's a scraper.

For a new site that isn't indexed yet, submit the sitemap, request indexing on the important URLs, and wait two weeks before taking drastic action. A self-referencing canonical tag plus real internal links pointing to the URL usually solves the issue.

Miguelrg72 said:

usually means the content got scraped and google indexed the copy first, on a new site that is not indexed yet is very easy to happen. self referencing canonical + get the page indexed fast (indexnow / api) before the scrapers. if it stays, a dmca on the copy is the only thing that moved it for me.

Click to expand...

In one of my sites, Google replaced my indexed page with that of an authority site. Although we can add canonical tag for html, it's not possible for docs & media. So, an alternative is to pass canonical tag through response headers that won't show up in content source but Google bot can decipher them and properly add canonical tags. However, it's difficult to do so if you have a lot of docs/media webpages without a standard URL structure.