Google webmaster reports non-existent links between syndicated sites

[prakash.sikchi]

We have run into an issue with linking that we are completely puzzled by. We syndicate our content to various clients, taking care to ensure that we have followed all the best practices that Google recommends for syndicating content. But recently, we noticed Google Webmaster report links from ClientA to ClientB, and we cannot figure out why it thinks that way. We have never created, and we have never found the links that Google Webmaster claims are there.

It is important for us to keep our clients isolated. Has anyone seen such behavior? Any ideas/pointers/hunches would be very much appreciated. Happy to provide more information.

We even asked on the Google Webmaster Forum (https://productforums.google.com/forum/#!topic/webmasters/QkGF7-HZHTY;context-place=forum/webmasters), but thought this might be a better place to get expert advice.

Thanks!

[RuthBurrReedy]

Is this something that is happening between just two specific client sites? Or is it more widespread for all of your clients?

[prakash.sikchi]

It happens with a few of the client sites. Google Webmasters thinks most of the other clients are pointing to these few.

[RuthBurrReedy]

Are you seeing this with any other link discovery tools, like AHREFs, Majestic, or Open Site Explorer? Or only in Google WMT?

[prakash.sikchi]

Only with Google WMT

[RuthBurrReedy]

It looks like the pages that you're referencing in the Webmaster Forums post do have some indexed duplicate content on them. It may be that Google is trying to figure out which version of the page is the canonical version, and then reporting all the duplicates as links back to the page that they deem the "original" page. I don't know why they would report this as a "link" when it clearly isn't a link in the traditional sense, but link reports in Search Console often contain some weird stuff that isn't really links.

What's clear is that Google understands that these pages are related, which makes sense since it's the same content, on pages with very similar design - the two subdomains are even using the same Google Analytics account. My assumption would be that the "links" are a byproduct of Google trying to figure out which of these pages are the duplicates and which should be treated as the original.

It may be that this is nothing to worry about - if you're happy with the pages' performance, I doubt you will see these "links" coming through anywhere besides Google Search Console. If you do want to keep the client sites more separate, I would recommend separating them out into their own Google Analytics accounts at the very least.