READ PART ONE HERE.
How quickly things change. In the first part of this series on social media and SEO, we dug into some of the reasons why search engines might give significant weight to signals from social media in ranking the search results.
Well just last week, Google and Bing confirmed what many SEOs have long thought to be the case – social signals from Twitter and Facebook do affect rankings. This great post at Search Engine Land is the first official confirmation from Google on the topic, and it is full of great insight.
This simple statement clarifies a lot of debate about the whether or not social signals impact web search. They do, plain and simple.
While this announcement still won’t convince everyone, the rest of us can move on to better understanding how authority works on social sites. What makes for an authoritative user, and how much value do they lend to links they share, or other social profiles they interact with? For marketers, the value lies in understanding this piece of the equation. These are also the details that search engines are likely to be much more protective of.
It’s all about authority
While we have little information from search engines about how they calculate user authority on social media sites, understanding PageRank (Google’s authority metric for web search) is probably a good first step to understanding how user authority works.
On social sites, authority is likely built and transferred in a similar way. A few important users following you or regularly engaging you builds your authority, just as a few links from important sites on the web can propel the authority of a page. Similarly, a groundswell of engagement from a large number of average users (provided the groundswell doesn’t look like it was manipulated) can also build authority.
So, how does authority “flow” through the social graph, and out to the rest of the web?
Followers and retweets are only the tip of the iceberg. There are a number of “interactions” that likely pass authority from user to user, and from users to links or shared content. We’ll look at this in the context of twitter, but these ideas apply to Facebook and other social networks too.
Quantity and quality of followers – This is the most obvious metric. Having a large number of non-spam followers, or a few highly authoritative followers can build authority for an account.
Your real reputation - If you are one of a couple twitter accounts linked from CNN.com, it’s easy for search engines to judge that you represent or are closely affiliated with CNN, and deserve some authority for that reason. It’s possible that this is the starting point for a lot of authority flowing through social sites.
Ratio and graph of followers – While this can be manipulated, search engines likely look at how a users’ follower base was initially built and how it has changed over time. Sudden spikes in low authority followers or wide fluctuations in the ratio of followees to followers could indicate manipulation, just as the quick acquisition of many low-quality links is known to be a red flag to search engines on the web.
Quantity and quality of mentions – Who replies to your tweets or regularly retweets your content? A large number of users sharing your content signals a wide-ranging influence. Having your content shared or retweeted by an authoritative user likely lends value both to your content and to your account.
Topical relevance - In general, how tightly themed are your tweets? While this in itself doesn’t indicate quality, it’s a desirable characteristic in the eyes of search engines (who need to match the content you are sharing up to specific search queries.)
Shared content quality - Bing has confirmed that they look at the quality of the links a user shares in grading users on social sites. Sharing quality content that other users are sharing – and being early to do so – is likely a positive indicator.
Inclusion in lists - Putting a user on a list is a way of one user vouching for another users’ knowledge of a particular topic. There are also a lot of relevancy signals in lists because users are often grouped by topic.
But this would be so easy to manipulate!
A lot of people seem to think that social signals are too easily gamed, and this announcement will result in an explosion of spam on social networking sites. I’m not entirely convinced. There is already an undercurrent of spam on Twitter and Facebook, and it’s possible that this announcement will lead to there being more. But it will be harder to drive real rankings with social spam than it is with link based spam, because it will be harder to derive authority on a social site than it is on the larger web.
Again, a comparison to PageRank helps here. Just as a link farm made up of thousands of low quality pages doesn’t generate a valuable link on the web, a manipulated social graph (thousands of bought or manipulated followers or retweets) probably won’t generate authority in itself. Why? There has to be an infusion of authority from somewhere – and in general, real users aren’t going to follow or interact with spam accounts.
What this means for SEO
This is an exciting announcement, and an exciting time to be an SEO. All of the sharing, connecting, and becoming “influential” that many forward-thinking SEOs and social media marketers have been advocating has been confirmed to impact rankings.
While an impact on search engine rankings is just one of the many benefits of using social media, it is the benefit most closely aligned with generating real revenue. For a lot of organizations, being able to see that link will mean that more resources and attention will be paid to social media.
This is also a compelling reason for SEO and social responsibilities to be more closely aligned. Social has value to SEO, and good SEOs who understand quality and authority on the web will have a valuable perspective on social media.