Google Authorship Oddness

Quite a while ago, Google launched a feature called authorship, which allows people who publish content online to link that content with their Google+ profile. The most obvious effect of this is that your Google+ profile picture and link appear in Google search engine ranks which, whilst it may or not benefit search engine rankings, probably makes your links more likely to be clicked on (I have no hard evidence of this other than the change in my own behaviour!)

Example of Google Authorship in action

I see this (and the implied data Google is collating in the process – there is already talk of an ‘author rank’) as having a real benefit for content creators so, as I am a big fan of Google+ anyway, I decided to give users of Social Hiking the ability to claim authorship of their maps.

The process of claiming authorship (when the author does not have an email address from the website domain) is:

  • the publisher adds a link to the author’s Google+ profile on the content page (with rel=”author” attribute in the link tag)
  • the author adds the publisher’s website to the ‘contributer to’ section of their Google+ profile
  • for profile details to appear in search results the author needs a recognisable headshot as a profile picture

In other words, the author approves the publisher to attribute content on that particular site on their behalf. In the case of Social Hiking, users add a link to their Google+ profile in the site settings, which is then automatically added to their maps (with the relevant attribute). Initially, no other pages had author attribution and the Social Hiking page was linked on all pages as the publisher (it is the same principle but with rel=”publisher” in the link).

Google infers authorship

Within a few days of switching the feature on, user profile images and links started appearing in Google search results next to their maps. Happy that the feature was working, I quickly forgot about it and moved onto other features…… until yesterday…. when I did a search for ‘social hiking” (as you do).

Example of incorrect Google authorship

Imagine my surprise to discover that Google had given authorship of the homepage of the entire site to one of the users (Social Hiking – Share Your Adventure is not by Chris Pilgrim!) The homepage had no authorship setup, so it seems that Google has taken the authorship of a page linked from the homepage and applied it to the homepage. Google infers authorship on pages without explicitly defined authorship! I have no idea if this is a bug or a feature, but it could potentially effect any site with multiple authors (and there are some more extreme examples).

For Social Hiking, I have been able to fix the problem by explicitly adding the Social Hiking Google+ profile as the author on all non map and user profile pages – whilst pages are not supported as authors (yet?), this seems to be enough to stop Google inferring authorship.

It seems the inferring goes further

When I added support for authorship on Social Hiking, I also set myself up as an author despite not having a suitable profile picture. As expected it did not work until I changed my profile picture last week, which triggered an email confirmation from Google.

Screenshot of authorship confirmation email

Oddly though, the example in the email was for this blog not for Social Hiking. This blog, which uses wordpress, has not been setup for authorship – there is a link to my Google+ profile, but it lacks the required author attribute. Also, whilst my blog is listed on my Google+ profile links section (actually now I come to check this it seems to have been renamed ‘other profiles’), it is not in the ‘contributor to’ section. Checking Google’s Structured Data Testing Tool confirms that I should not be the author of my blog in the eyes of Google…..

Result of the Structured Data Testing Tool for this blog

yet I am…..

Authorship for my blogI can only assume that Google has seen that the blog links to my Google+ profile, which in turn links back to the blog, and decided that is sufficient indication of authorship (of course it can only help that I have a vanity domain for this blog!)


As I was writing this blog post, I noticed that Dean Read’s Google+ profile was linked to his Audioboo profile in Google search results – this surprised me, as I know that Audioboo does not have any ability to link your Google+ profile.

Taking a look at the source code shows that Audioboo uses rel=”me” (albeit with a nofollow) on a link to Dean’s blog, which is setup with Dean as author. This connection seems to be sufficient for Google to infer authorship (Dean has a link to his Audioboo account from Google+, which is presumably sufficient to confirm this connection).

Want to know more?

There are plenty of resources out there, but a good place to start is the Google Authorship & Author Rank Google+ community.

For more about why authorship and authority on Google+ are so important, have a read of ‘8 reasons why you need to establish authority on Google+’