Agent Rank As Google Building an Identity Service Based on Authority?
What does it mean to call Agent Rank an Identity Service? Might that have implications for how web pages rank at Google? If Google starts using authority signals into those rankings, it very well could.
At the Edinburgh Intl TV Festival on August 28th, 2011, at a Q&A with Eric Schmidt, Andy Carvin from NPR asked about Google’s insistence on using people’s real names. He received a long response that pointed out using Google Plus as an identity service using a ranking signal built into it.
But my general rule is people have a lot of free time and people on the Internet, some people do really really evil and wrong things on the Internet, and it would be useful if we had a strong identity so we could weed them out. I’m not suggesting eliminating them. What I’m suggesting is if we knew their identity was accurate, we could rank them. Think of them as an identity rank.
The Return of Agent Rank and Portable Digital Signatures
I recall a Google patent describing the use of digital identity on Google’s patent filing involving Agent Rank. I wrote about it in Google’s Agent Rank / Author Rank Patent Filing.
The patent describes how authors can mark content published on the Web with a digital signature. That would be regardless of it being a web page, article, blog post, or even a comment. It also tells us that authorship might influence the rankings of that content by associating a reputation score with it.
The patent also describes ways of using metadata to provide information about whether published elsewhere. This would be somewhat like the syndication metadata tags developed for Google News. (See also, Google on Duplicate Content Filtering and News Attribution Metatags.)
An Update to the Agent Rank Patent
This summer, Google updated the reputation scoring system described in the Agent Rank patent. That makes it more likely to use an authorship markup approach and a system like Google Plus. It’s not the first continuation of the patent. An earlier continuation was published in 2009.
Both continuations of the original patent contain substantially the same description, but the newest version of the Agent Rank patent adds an interesting element in its claims.
Here is the first of the claims listed in the agent rank patent filing:
1. A computer-implemented method comprising: evaluating a document that is hosted on a site, the document including a content item to which a maker of the content item has applied a digital signature; determining whether the digital signature is portable; if the digital signature is portable, using a reputation score associated with the maker in calculating a quality score for the document; and if the digital signature is not portable, using the reputation score associated with the maker in calculating the quality score for the document only if the digital signature is fixed to the site.
Note the use of the word “portable.”
That seems to be one of the more significant updates to the patent. I think it makes it more clear that associating a Google profile with content authored by someone is like applying a digital signature to that content.
Note that someone doesn’t have to use Google Plus with authorship markup since they need only link their profile page with their Google Account under the processes described by Google on their help page for using the markup.
Reputation Scoring under Agent Rank
The Google+ authorship markup can identify who the author and possibly originator of content might be. The Agent Rank process also involves adding a quality score to a document based upon the reputation score of its author.
It’s quite possible that Google might use information that it finds on the Web and at Google plus that it can clearly associate with specific authors. This would create a credential score of the type that I described in the post How Google Might Rank User Generated Web Content in Google + and Other Social Networks.
Google might also issue Authorship Badges as well that we could use in other places such as comments on blog posts, or upon guest posts on blogs where we might not have an author profile page, or upon a site where we might submit articles.
Google’s Panda updates emphasized the “quality” of the content found on the Web. An authorship reputation could easily be another signal of quality in ranking pages.
Agent Rank Conclusion
As I noted in my last post, Google’s New Freshness Update: Social Media Has Changed the Expectations of Searchers, there are many reasons why Google might look more at social signals, including the ability to surface fresher content faster.
If Google is to do that, it needs signals other than PageRank to find very fresh content. That is because PageRank tends to work best for content that has had time to attract some links. You can add an author’s reputation score based upon contributions to social networks. Or through very recently published content and meaningful interactions with others in social networks and comments. That will add an element that doesn’t rely upon a link graph the way that PageRank does.
Adding authorship markup to the content that you create would be a step in this process. It would enable Google to associate a digital signature to content that you might create at many places on the Web.
Google has taken steps to install authorship markup on its own. In a June post about Authorship Markup on the Google Webmaster Central blog, they tell us:
We wanted to make sure the markup was as easy to implement as possible. To that end, we’ve already worked with several sites to markup their pages, including The New York Times, The Washington Post, CNET, Entertainment Weekly, The New Yorker, and others. In addition, we’ve taken the extra step to add this markup to everything hosted by YouTube and Blogger. In the future, both platforms will automatically include this markup when you publish content.
You May Be Using Content Google Knows that You are the Author of
You can publish content to Youtube or Google Blogger, or one of Google’s organizations. If you do, you’re already using authorship markup.
Some other recent posts elsewhere that discuss Agent Rank and Google Authorship markup:
- Jim Boykin at Internet Marketing Ninjas – Google Agent Rank and Reputational Scores ¦ It’s About Content and Writers and Panda!
- Simon Penson at SEOGadget – Google’s Heading for Life after Link Trust – Here’s How to Prepare
- John Doherty at johndoherty.com – How A Search Engine Might Determine Author Authority for Rankings
- John Doherty at SEOmoz – Social Network Spam and Author/Agent Rank
How do you feel about Google Plus as an identity service?
Will you be adding authorship markup to your pages if you haven’t already?
Interesting. I think this has real potential for Google – being able to group content online by who created it, especially if they can link it to a Google+ account.
Spammers could still create fake identities (it would just take a lot of work). I think this would just give Google one more layer of data to use.
The idea of ranking based on authorship makes sense to me. Using Google+ to define that authorship seems a bit lame to me, but I can see that being done. Page rank can pretty quickly get some value due to authorship (if the author chooses). If an author has authority (according to Google) and they link to their own new content they will get pagerank (and compared to someone with no authority that will be a big difference). Also Google already seems to have significant boosts to new content (in addition to liking old content :-).
Similarly not using Twitter… indications is lame. Google should be able to get value from “no-follow” links. The idea of no-follow I think is just not very good. Why through away an indicator, even if you have less confidence in it. Google should be able to get meaningful signals from things like authority on topics adding links to new content (even if no-follow) which can help deal with time needed to build up page rank.
Great discussion.
We should also think about the potential of abusing the system and the cost of quality content will go up. I would say changes are good but the way google is makeing all the rapid changes its driving me mad.
– livingseolife
the Traffic on Google + more than on facebook, this news really busted me up that a new social media takes place. I don’t know its completely true or its just fake. But its kinda attracted news for everyone.
I personlly feel this is not a great idea. Spammers can open several accounts to promote their website using Google+. Also any one can ask friends and family members and can get their favour. Its only my persionally opinion but Bill tell me what Do you think?
Google will eventually integrate its +1 into the search algorithm itself so human votes will have an impact on search ranking.
I don’t know what to think about this… It has some advantages but…possible danger is that it will create an impression of “true authority” which is not exactly truth because it can be misused…people will just be more creative in creating ways to misuse it and people who are sincere may be disadvantaged because they are not interested in finding out how it works in details…I don’t know…but some arguments that are mentioned makes sense…
Hi Adam,
Definitely one of the things that a search engine needs to consider when it decides what kind of ranking signals that it will use is what the cost of attack is on those signals. With an agent rank or authorship rank approach, a lot of time would definitely be needed to be invested by someone trying to use social signals to affect rankings. The cost of attack on them is fairly high because of that, which can make those signals more reliable.
The fact that Google could more easily identify when identities are fake, and ignore them when they don’t contribute much to Google Plus or interact on the social network in meaningful ways makes using social signals more valuable. Google tying the Google Plus signals to Google Accounts means that Google has a lot of access to collateral information about how a person uses those networks – much more than looking at Twitter of Facebook where Google doesn’t have access to information such as the IP address where someone might have tweeted from, or updated their status from.
Hi John,
Using social signals in ranking pages based upon reputation scores for the authors of that content means that fresher newer material that might not have had time to attract links because it is so new still has a chance of ranking in search results, in social search results, and in real-time search results as well.
I’m not sure that a nofollow approach from Twitter really matters, if Google has the ability to use social networks to identify recency-sensitive topics and pages being linked to on those topics from Twitter and Google Plus. If Google were to rely solely upon links from places like twitter, they would have to do some very fast calculations and iterations of PageRank on a regular basis, and that just might not be the best use of their resources, especially if they had to also try to find a way to filter out attempts to manipulate search results from people engaging in those social networks that have created sock puppets or fake accounts solely for those purposes. Filtering out those social network profiles first makes more sense.
Hi livingseolife,
I am wondering how all of the changes over the past year or so at Google might be perceived by people who aren’t following them closely, but start seeing a lot of changes such as the appearance and then disappearance of real time search, the Google Authorship profile images showing up in results, plus buttons, links to block sites, and more.
A lot of these changes do seem to be happening very rapidly, but for a number of them it appears that Google has been looking at them and experimenting with them for a good amount of time, and in some cases a number of years.
Hi Gary,
There have been a number of reports on tech blogs about the traffic levels to Google Plus, and how it might seem like there isn’t much, at least publicly. But there does seem to be a lot of activity shared between people on different circles that is below the surface, and we’ve been hearing that from people at Google.
Google’s project manager in charge of Google Plus recently discussed this on Bloomberg TV, responding to charges from Facebook that Google isn’t being used much with a statement that he’s glad to being underestimated by their competition:
https://plus.google.com/113116318008017777871/posts/4Sq4NWod297
Hi Tevin,
The point behind Google using something like an author rank is that it enables them to increase the amount of data and information that they have to rank pages and metablog data like social signals based upon information they collect about the reputation of people who post and interact at places like Google Plus, and it allows them to verify the authorship of content found on the Web.
If their friends and family members are active on the Web, contribute and interact on social networks and on the web in meaningful ways, then Google has more information to use to avoid manipulations by people who are only interested in spamming and manipulating search results.
Hi Mina,
There is a risk that the kinds of authority and reputation scoring that Google might do has a potential to mislead, but one of the things that Google seems to have baked into their approach is the ability for people to go to the Google Plus accounts of people who are shown to have Google profile images associated with content they have authored, and judge for themselves about the authority of those people.
Hi Janice,
I suspect that not every +1 will have the same value as other +1’s, based upon who is doing the plussing.
given the focus and attention google is giving to g+, more and more g+ would be integrated in to all their products. and the social element will influence everything – from personalization to search results.
Hi Pradeep,
There have been a lot of people writing about Google Plus not being a social network, but rather a growing platform that will connect more and more of the services that Google provides together. I’m not sure that they will tie all services that they provide to Google Plus, but it does seem that there is a pretty strong move towards every service that requires a login to use a Google Account, and an increased integration of services.
It makes a lot of sense for Google to do that, and the idea of using a specific account with a digital signature unique to that account makes it easier for Google to provide services to users that take advantage of those connections between services. It can also help Google identify fake profiles, who originated content that might be duplicated elsewhere on the Web and more.
I agree with most of the comments here. I have a question. I have been told by the IM ‘honchos’ that having the Google Plus icon on my websites will not help Google to like my site any more than if I was not ‘google plused’. Has anyone done any testing to see if that helps your SEO??
Hi Lanny,
There are a lot of Internet Marketing “honchos” in the world. 🙂
We don’t exactly know the role that having a Google Plus icon on your site might play in SEO since it’s all very new, and in many ways a moving target. At this point it does appear that Google has at least a rough outline for how such things might play a role in SEO, but it’s not one that they’ve explicitly written about or published about online, and anyone who tells you that they know for certain whether or not those types of icons will help you or not really doesn’t know.
Great blog Bill yeah it’s a massive grey area at the minute. I think if Google give more credence to +1’s than Facebook likes it will leave it wide open for spammers with multiple gmail accounts to manipulate.
Hi Sean,
Google can track spammers with Google/Gmail accounts, and ignore +1s from fake profiles, sock puppets, and those who really haven’t been very active. They can’t do that with Facebook Likes.
Nice link bait from SEO Moz