Social networks refers to collections of entities (people, companies, etc.) grouped by interests, by demographics, and by many other categories, where each entity has a relationship with the other entities in that group. This means that a social networks is a very interesting source of data and can be very lucrative for the companies who own it.
With this in mind, it's clear why access to this information isn't freely given to third parties. In fact, the majority of the resources found on a social network website's domain (e.g. Facebook.com) aren't analyzable by crawlers (software used to analyze and scan a domain), and aren't included in search engine indexes. Now let's introduce another definition that has been gaining ground in conventional media: the DeepWeb, or the part of the Internet that's hidden, submerged. In practice, this refers to all of the resources which aren't indexed by search engines, and aren't accessible by most web users. This includes information from social networks.
If the majority of a social network's information isn't visible, what is the point of integrating these networks with search engines? After all, the information used to define organic positioning would be based on incomplete data.
This isn't to say that information from social networks that is indexed by search engines isn't taken into consideration; this information is actually given the same weight as articles, blog comments, forum threads, etc.
How does Google value the influence of social networks when calculating search result rankings?
In 2014, Matt Cuts posted a video (YouTube Video) in which he explained why it didn't make sense to include signals from the two largest social networks of the time, Facebook and Twitter, into Google's ranking algorithms (the calculation the search engine assigns points to a resource). Later, Google signed a partnership with Twitter to analyze tweets, but abandoned this project.
In a 2015 Hangout hosted by John Mueller (YouTube Video, minute 19:55) the relationship between search engines and social networks was explained once more. There is no special relationship between them, and he claims that no direct signals are used to calculate a resource's search rankings.
How Google Works
In this paragraph, we'll explain the building blocks of a search engine in more detail. This should clear up the reasons for all this confusion about whether social networks are important for organic positioning, or whether they don't have a special relationship with search engines, and are treated the same as other web resources.
Google has released a few patents in which we find a distinction between two important structures that exploit its system for organizing systems into graphs. A graph is a means of representing complex relationships for a specific domain. Search engines use the Link Graph and the Social Graph. The Link Graph is used to keep track of links between the different websites that make up the web. This graph is shaped by a website's internal and external links.
The Social Graph keeps track of mentions, citations, and co-citations. How many people are talking about your brand online? How “famous” are you online? How many domains contain a link to your domain? This last question shows the intersection between the Link Graph and the Social Graph.
What are mentions, citations, and co-citations? And most importantly, how can you create them?
Mentions are citations of your brand or domain within a resource. A mention isn't a link, but rather text within an article, forum thread, blog comment, or public Facebook post, for example, as long as this page is present and archived within a search engine index.
Citations are links to an external domain with the intention of creating a connection.
Co-citations are also links, but they are defined as such when several links to different external domains are located within a text, dealing with the same topic. For example, if an article comparing two or more cameras includes links to various online stores, those are co-citations.
Together, these three elements are referred to as the Social Signal. They're not the social signals many picture, rather they work more like “indirect influence” John Mueller refers to in his hangout. Either way, if these elements are used well, with a good balance of the three, they can form the basis for an effective strategy for earning links from other domains.
A closer look at Google's 2016 patent
In its patent, “Searching Content of Prominent Users in Social Network”, which you can read here link Google tries to define a method for extracting social connections contained in a search engine's index for a given brand and/or its domain, if they do coincide.
Here's a summary of some of the steps:
- Receiving the user query
- Collecting information and organizing connections to that brand/domain
- Refining the results based on connections in the search engine index.
- Visualizing search results which are also weighted based on social signal.
Later on in the patent, we also find interesting passages concerning the filtering of social connections found by Google's crawlers, analyzing the WEB, information that is considered important, and how to organize it. If you're interested in learning more about this, read sections 0007, 0026, 0031, 00041 e 0048.
In this article, we've explained what a Social Signal is, and how there is no direct or special link between organic positioning and your activities on social networks like Facebook. This doesn't mean that it is useless when it comes to positioning strategies, just keep these warnings in mind when using it.