I have long believed, but been unable to prove, that social signals support organic rank, making the repeatedly cited correlations between social and organic rank in Moz’s 2011 and 2013 ranking factors analyses both intriguing and maddening. In the absence of a clear causal mechanism, the correlation leaves my most important question as a marketing strategist unanswered: how much time, effort and money do I allocate to social and organic respectively, and what result should I expect to get back? Is social a brand and engagement vehicle with some indirect impact on numbers of new visitors, is it an essential part of organic search visibility, or is it both?
I decided to explore a simple explanation that argues it is both, and this post is my attempt to make the case and explain how it should affect social and organic strategy.
I first considered and rejected a number of the more obvious theories discussed in the comments to the earlier Moz Blog post:
- Direct causal relationship: Google reads social activity and includes it as a ranking factor, at least as it relates to Google+. Google has denied this, so unless you’re a conspiracy theorist, this is a non-starter. Mark Traphagen’s finding that profiles and pages in Google+ have page rank and pass link equity is compelling, but the bulk of social activity still happens on Facebook, i.e.: a registration wall and a whole lot of nofollow tags away from the prying eyes of Google’s crawls.
- Common causal relationship: some other independent factor (perhaps a strong brand) drives both social activity and rank, creating correlation but not causation. This is undoubtedly true to some degree, but it is also as vague as it is self-evident. Of course, you want a strong brand, and a strong brand no doubt helps with inbound links and other indicators of authority, but the question I’m interested in is how much investment to make – where and when – to accomplish what. This theory does not answer those questions, since you can’t index something a brand metric (say, recall) to a specific organic rank.
- Indirect causal relationship: social activity drives some other action, such as inbound linking, that in turn affects rank.
The third is the most compelling, but I know from personal experience that there does not have to be any additional activity for something to rank. I saw a specific case of this earlier this year when a blog post I contributed to a corporate website with a domain authority of just 37 ended up ranking in third position on a fairly high-profile search term, ahead of Huffington Post and just behind a prestigious print publication with a strong online presence. The post was promoted solely through the company’s social media presence. The article received more than 150 likes, four tweets, just three +1’s (one of which was from the company’s own Google+ page), and not a single external link.
How is that ranking possible, unless social activity directly drove the rank? This brings me to a fourth theory:
- Social activity creates user behavior that gives Google an understanding of the relative quality of content on third party websites, an understanding it then uses to order search results.
- Information that Google receives when you use Chrome is processed in order to operate and improve Chrome and other Google services.”
Google doesn’t collect general browsing data by default. However, it does if you sign in to Chrome:
Google has been increasingly touting this feature of Chrome with TV ads that talk about the fact that you can move from one device to another with a single browser and pick right back up where you left off on the last one.
To confirm what is and is not passed back and forth between Chrome and Google, I monitored all http traffic sent to and from Chrome as I browsed the web. When I was logged in, every request I entered for a URL generated two types of entries: multiple commands to complete the search result as Google treated my typing as a query and then a Chrome sync operation with details on the URL visited. Clicks on links also produced the same sync operation. I then logged out and performed the same actions, and as expected, the search complete commands continued, but the sync operations did not.
With the number of people using Chrome and Android, this gives Google a significant body of data about time on page, time on site, and bounce rate. Is there any reason to think Google doesn’t mine the resulting aggregate data to understand how long someone spends browsing a page they are referred to from social media? It might seem intrusive, but Google does far more intrusive things, like serving ads based on the content of emails and using omnibox data (which are not immediately made anonymous) to improve its suggestions service.
Assuming that Google uses the data as I’m suggesting they do, the implication is obvious: building a community via social media is a critical part of an SEO strategy, because it makes it easier for you to put your content in front of users (users who happen to be predisposed to engaging with your content in a positive way) and give Google insight into the quality of that content. If “all” you do is gather your existing audience into a cohesive community, build good content, and promote that content to your community, rank will – more or less – logically follow.
If you use Google+ as one of those distribution channels, you’re increasing the likelihood of putting the content in front of users who are logged in to Chrome, hence the strong correlation between +1’s and rank. Consider conducting hangouts if necessary to drive your audience to Google+, if they’re not already there. Consider paying special attention, too, to the engagement metrics for users who are accessing your site via Android – if they’re not strong, figure out why and address it, since those users are another segment likely to be logged in to Chrome.
Even if I’m right, this doesn’t provide any neat, simple tricks for ranking. It merely provides greater confidence in making the investment in the hard work behind these activities. It also means that content is a double-edged sword: if you put out bad content and promote it via social channels, it can harm your rankings just as quickly as it can help them.