Social Signals Suck: Why Google Still Prefers The Gameable Link

Aren’t you tired of constantly hearing about how Google is going to replace traditional links with social links soon? Every few months, an “expert” publishes a case study explaining that social links are ranking factors, and then surmises that social media will become the new SEO in a couple years. While this is a great strategy for the expert to get publicity and links to his/her blog, the unfortunate thing is that no one outside of Mountain View can even come close to making this call.

Consider this – no one actually has access to Google’s algorithm or the ridiculous amount of data they have collected on websites, social platforms, authorship, etc. In fact, about 5 years ago, Matt Cutts explained that Google has over 200 signals and that many of them have more than 50 variations within a single factor. And that was 5 years ago! Now imagine what that algorithm looks like now?

One famous quote from Albert Einstein comes to mind:

“The more I learn, the more I realize how much I don’t know.”

The point is that there are several tell tale signs that indicate that what webmasters see is simply correlation and not causation. Here are a few reasons Google’ won’t be forgetting about links any time soon.

Lack of Access

Facebook is the largest social media platform in the world. Because both companies thrive on advertising, Facebook views Google as a fierce competitor for advertisers’ budgets. To maintain a competitive advantage when it comes to social advertising, Facebook has never given Google access to crawl its ecosystem. While Google does have access to the number of likes and shares as well as names of users and pages, Facebook doesn’t allow Google to get much else.

This poses a problem for Google – how do you qualify the validity of user accounts, the relevance of a “Like”, the authority behind a “Share”, or a dozen other qualifying factors if you don’t have access to more than just a name? After all, Google’s isn’t the type of technology company to simply rely on the number of “Likes” – an easily manipulated figure – as a ranking signal, is it?

Without enough user background information, Google cannot give much weight to social signals.

Getting A Social Link Is Too Easy

As Google has always told the SEO community, the truest and most valuable link is the contextual editorial link. Not the blogroll, footer, or widget link. But the one you get from the Wall Street Journal, New York Times, TIME, BBC News, etc.

One of the many reasons the editorial link carries so much weight is that it requires effort to give it. Someone, somewhere had to thoughtfully create something (e.g. text, sound, video, and/or graphic) and reference this link within the content.

Guest posting had a meteoric rise because it addressed this issue. Low-quality guest posting was and high-quality guest posting continues to be a great trade off – contributor produces content and gets a link (who are we kidding? it’s always partially about the link), while webmaster gets free content and ideally promotion from contributor after publication.

Even without guest posting, getting an influencer to tweet or share your content is a hundred times easier than getting a link or guest posting opportunity.

On the other hand, social linking and sharing sometimes amounts to nothing but a popularity contest. Generally speaking, celebrities have the most followers and influence on social media. Does that mean Kim Kardashian tweeting about a diet pill and getting 10,000 RTs for the company’s product page should result in Google ranking the product page #1 for a search on “best diet pills for women”?

Similarly, should Kim Kardashian, who has 35.3M followers without an authority blog, be a social signal of value when the Food and Drug Administration’s own Twitter account only boasts 111K followers? A comparison of their engagement statistics makes a case for social signals as significant ranking factors even slimmer, although the FDA could use a guide on how and when to post on social media.

Why We See Correlation and Think Causation

As I mentioned at the beginning, without having full access to the list of factors Google’s algorithm evaluates, it’s impossible to understand why we see the results we do. Statements attempting to explain the direct causal relationship between social sharing and rankings are akin to stating that lions don’t live in Los Angeles and therefore lions must not like Californians. If I interviewed 100,000 Californians, I’m sure 99,999 (because we all know one crazy Californian probably keeps a lion as a pet) will also confess that lions don’t live with them either, so my hypothesis must be accurate.

However, in the case of social media and rankings, isn’t it more likely that bloggers and companies with large social media followings also have authoritative blogs, “influencer” or “expert” statuses within their communities, and relationships with other authority bloggers as well? Content from these bloggers would rank well regardless of social sharing.

Another possibility to consider is that, although social sharing does not have a direct effect on links, SEO and search engine rankings, it does build exposure and awareness. That reach and awareness could translate into a link when a blogger comes across a Tweet promoting a valuable piece of content and the blogger decides to share it with his/her audience in an upcoming post. Without the Tweet, the traditional editorial link would not exist.

Final Word

The point of this type of discussion isn’t to discourage anyone from using social media or, at the other extreme, only focus on building traditional types of links. As a digital marketer, you must take a holistic approach to building your web presence. The fact is that the best links are always natural because you have no idea how you exactly got them, but all that’s important is that you did. And that’s because traditional links are still the backbone of organic search engine rankings.

^{* Lead image adapte from Joe The Goat Farmer}