Does Social Media Affect SEO?
In this article we talk about the reasons there's so much confusion around this topic and answer the question.

In this article we talk about the reasons there's so much confusion around this topic and answer the question.
There's a good bit of confusion about whether social signals factor into Google's ranking algorithms. There are articles on both sides from reputable sources and a whole lot of hearsay and conjecture.
I originally wrote this article by say, "No, social media does not impact SEO in any way, and here are all the reasons why…"
But that was the wrong approach and was a bit dismissive of the intricacies of marketing.
The real answer is that social media does impact SEO, but not in the way you might think. The following is an overview of the source of all the confusion and a clear answer to the question.
I think the confusion over whether social signals impact search rankings stems from 3 primary sources:
In Dec, 2010 Matt Cutts released the following video in which he very clearly states that Google is using social signals in the ranking algorithms and that they're working to take social reputation / rank into account as well.
Then, in Jan, 2014, he release this video seemingly stating the opposite.
All the commentary on this new video suggests Cutts clearly stated that social signals are not a part of the ranking algorithm. I don't agree that it's quite that clear cut. In the very first part of the video he states that social pages are treated like any other page for the purposes of crawling and indexing, but he goes on to say,
"as far as doing special, specific work to sort of say, oh you have this many followers on Twitter or this many likes on Facebook, to the best of my knowledge we don't currently have any signals like that in our web search ranking algorithm."
This comment seems to me to be more related to assigning certain social accounts an authority score than not using social signals at all. Cutts then goes on to explain that the search engineers would be hesitant to use information from social profiles because they could be blocked from accessing those profiles again in the future to get up-to-date information.
He uses an example of a wife and abusive husband who are no longer together and suggests the engineers wouldn't want to extract data that they were married, use that data somewhere, be blocked from accessing the page for whatever reason in the future, and not know when they got a divorce - in other words, act on incorrect, out-dated information.
So Google is worried about using inaccurate information. Great. That's always been a problem and it's not limited to social sites. He then says (emphasis mine),
"don't necessarily assume that because there's a signal on Facebook or Twitter that Google is able to access that."
Social pages might be blocked or there might be "nofollow" on links, but how is that different from the rest of the (non-social) web?
He then says that content with lots of likes on Facebook isn't ranking because of those likes, rather it's ranking because it's awesome content earning real web mentions and links, and because it's awesome it then gets liked on Facebook (correlation vs. causation).
Finally, he wraps up the video by saying (grossly paraphrased) that maybe at some point they'll be able to imply authorship with some accuracy, but right now it's just too difficult.
I think these last two points are the reason all the commentary suggests that no, Google does not use social signals at all in it's ranking algorithms. But most of the examples Cutts uses in the video are specifically related to authorship and applying authority to certain social accounts. That doesn't answer the question of why they can't use Facebook likes as a ranking signal in any way.
Okay, so that long-winded diatribe is the first reason I think there's a lot of confusion over whether Google uses social signals in it's ranking algorithms. On to the second reason…
It's easy to look at two data points that are strongly correlated and assume that one of those data points actually caused the other. Most of the observations that suggest social signals influence rankings are incorrectly making this assumption.
People don't understand social media. They hear, "oh you should have a Facebook page for your business" and all this hype about social but there's very little understanding of how social media is supposed to be used or when and if it's appropriate. I think the result is a tendency to believe that social media for social media's sake will solve all your problems, including search rankings.
In 2010, Google probably had every intention of using social signals, and probably still does, but with the expiration of their data feed agreement with Twitter in 2011, they probably realized it may not be a good idea to rely on social networks too heavily, if at all. At least not at this stage in the game, and that's reflected in the 2014 video.
I don't think Google wants to rely on a handful of very large 3rd parties for major aspects of their ranking algorithms.
Google might devote serious resources toward integrating Facebook signals, for example, into the algorithm and then one day Mark Zuckerberg could wake up and decide he didn't want Google crawling Facebook, flip a switch, and Google loses all that valuable data that their algorithm relies on – gross oversimplification, I know, but you get the point.
I think the 2014 video above is a reflection of the lessons Google learned when the Twitter deal expired. Facebook pages are owned by Facebook and at the whim of its decisions.
Ownership of individual web pages, on the other hand, is far more distributed and those pages are far less likely to suddenly go offline or be blocked in bulk. Likewise, Google can use signals from its own properties without fear they'll lose access to that information. Because of this I don't think Google is relying on social signals, nor will they for the foreseeable future.
Neil Patel's article is perhaps the most influential of these – Neil's a well-known guy in the industry and his article is one of the first you find when researching the topic.
But it boils down to what Matt Cutts said in the 2014 video. These observations are correlational and social signals are not causing ranking changes. Rather, what's actually happening is the better your content, the more people link to it, view it, share it, search for it, etc. More Facebook likes can definitely lead to more views, links, shares, etc, but they aren't directly causing the increase in rankings.
If it happens, I think it's going to be a while. I'm sure Google wants to do this, but as I mentioned earlier, I don't think Google wants to rely on a handful of very large 3rd parties for major aspects of their ranking algorithms.
I suppose the answer to that question is, it depends on what you mean by "using social media." If you mean should you create a Facebook page for your business and spam all your friends telling them to like it in the hopes that you'll score some of that internet money, then no, you shouldn't be using social media.
Think of social media like you think of networking. You don't walk into a networking event and say, "hey everyone, I just started my business, tell me you like it!" That would just be weird. Instead, you walk into a networking event and give a really useful presentation, or you talk with someone about how you solved a problem similar to one they're having. In other words, all the networking does is facilitate the sharing of your already useful information.
Social media is the same thing as networking. It's not the end in itself, it's just the channel through which you share your useful information.
SEO has evolved quite a bit over the last few years. Google now takes into account many factors outside of what most people consider SEO. Brand awareness, greater association of a brand with a particular product or category, and news are just a few of many things that can have an effect on search rankings.
If you run a series of TV commercials promoting your new "Super Product Alpha," people are likely going to start Googling your company name + "super product alpha." The result is Google starts to develop a picture of your company name being associated with "Super Product Alpha," making it much easier for you to rank for that term. This is brand awareness impacting search rankings indirectly.
The same thing happens when you increase brand awareness through social media (or any channel for that matter).
Aside from brand awareness, greater reach of your content can indirectly influence search rankings. This can be a result of a TV or radio campaign, online display ads, or social media.
AJ Kohn explains this topic well in this post. As an example, the process might look like this:
So, in this case, social media is the channel that kicks off the sharing frenzy and gets your awesome content in front of more people, but it isn't the social signals themselves that influence your search rankings.
So it should be clear now that social signals, such as likes, friend count, shares, etc, don't directly impact SEO.
However, social is just another marketing channel. And you can use any marketing channel, social included, to promote your content and brand in a way that increases brand awareness and exposure to your content. The result of this can be improvements in the things that do influence SEO directly, such as backlinks and brand-associated searches.
Some excellent articles on both sides of the issue: