How Relevant are Social Signals for SEO?
Most social media sites consequently use the nofollow attribute, and this would mean that the traditional approach to ranking pages is not an initiator. Of course, there are those who believe that Google is selective in the sense of treating nofollow links.
How Important are the Social Signals in Search Engine Ranking
Personalization and Social Media
One area that is certainly worth consideration is personalization. If I am logged into Google or Yahoo, the social elements certainly change the manner of ranking results and the presentation of the query. It is happening because of our established connections in the social domain. What I am going to write in this article refers to more non-personalized results.
Signal and Ranking Factor
One should keep in mind that the signal is not always the same as the ranking factor. Google said: “We are using it as a signal in ranking of organic news and results. Also, we use it for improving news on the basis of how many people shared an article.” Other types of signals are:
- The concept of trust (sites and users)
- Temporal (speed)
- Context (semantics)
Most sensitive are Discovery and speed. Certainly it can be said that page cannot be ranked if it is not indexed so that social channels can have as much effect depending on how much is invested in discovery. There are elements Such as the QDF (query deserves Freshness). The element of speed with which we share content might play a leading role, but I’m not sure it deserves great attention.
When it comes to the social signals that play a role in the rankings, I would mention:
- Link sharing (by email, instant messengers, etc.), Or bookmarking content for someone else.
- Posting content (e.g. Adding to the blog where it becomes visible to others)
- Applying tags to the content visible to others.
- Achieving sales over the webpage.
- Promoting or pulling content in the results.
- Posting comments on news.
Authority and Trust
Entity that shares the content is supposed to be a reliable source for each search engine that could be used as a non-personalized non-ranking factor. That is the meaning of the signals because social networks can be inherently noisy and behave as spam.
In determining different weights – based on known data entity, the lower part of the spectrum will have a much smaller impact on ranking or it will not be at all, and this can help with spam where it appears excessive.
We can say that the quality of your interactions, the value of your contributions and content you share are observed and analyzed in decision making process. Building a reputation is considerable effort and it is still too easy to be lost. So always consider your concept of ranking based on trust and strictly separate links to good content of links with bad content.
Correlation and Causation
Recently a small group of people carried out an experiment that proves that social activity is in a good correlation with rankings of known search engines. The study included a range of topics: a huge number of keywords, SERP’s, titles, millions of Facebook comments, billions of likes, etc.
SEO experts, however, still advise to be careful because it is difficult to understand and embrace all the factors involved in ranking and sometimes everything just seems and looks so accidentally.
What is the real attitude of Microsoft, Google and Yahoo?
They generally recognize that they take into account the reputation of the author and that his authoritarianism is independent of the ranking, but it is currently being used in limited situations in ordinary web search.
What is this supposed to mean? So certainly that either Google or Bing does not know the value of such signals and if they are themselves uncertain about the value of their own measurements, then we can only assume how they “look” at third-grade Facebook, Twitter, signals.
Point of this article is to encourage you to think and to make your own conclusion. When we talk about the big dark giants, such as Google, Yahoo, Bing we can never reliably know what is happening.
Ranking based on tweets, likes and pluses make little sense. Which one is of bigger value Share or Like? What difference would it make? Remember that before +1 existed Buzz? Remember Twitter firehose that is now turned off? Well not really, just got replaced with new service called Pinerest!
It makes sense to look back to user entities within social networks instead of the actual sites where users are active.
Whether they should suspend all SEO activities and go exclusively with spamming via social networks and links? It would not be a good idea for sure.