Search Engine Optimization is not an exact science. Figuring out the lengthy, ever evolving, and sometimes confusing, mix of traditional “on-page” strategies, “off-page” factors such as link building, and newer influences such as social media is an on-going challenge.
It is obvious that social media is going to be playing a greater role in search engine results in the future, but questions of how soon it will happen, how significant its influence will be, and what will actually be measured is still up for debate.
Search Metrics recently released results of a Search based study that indicates that the move to “social media” influenced search results may be happening faster and with greater influence than I would have guessed.
The study, completed in February and March of 2012, consisted of analysis of “10,000 selected top-keywords, 300,000 websites and millions of links, shares and tweets”. The results were shared as correlation coefficients.
The 5 factors that had the greatest positive correlation coefficient to search results were:
#5 Traditional SEO “On-page” Strategies: The most relevant of these was having the Keyword in the Domain name. With a lesser, but still positive correlation coefficient, was Image Count, Keyword use in the URL, and Keyword use within the description tag.
#4 Tweets: The study does not clarify whether the correlation coefficient relates to the number of tweets, re-tweets, or other usage data, but does indicate that Twitter had a fairly high correlation coefficient.
#3 Number of Backlinks: As expected, backlinks are shown to be high level influencers of search engine results pages. It also shows that a wide variety of backlinks, including no-follow links, influence search results.
#2 Facebook: The number of Facebook Shares, Comments, and Likes all have positive correlation coefficients to search engine results. Facebook Shares was actually officially the number one factor in the study.
#1 The unofficial number one Search Engine Results factor according to this study is, not surprisingly . . . Google +1’s with a correlation of .41. Search metrics mentions this in the report, but dooes not include it in the official results.
Additional Results of Interest:
Brand Power: The big brands like Sony and Nike are not held to the same on-page SEO requirements as other websites in that they often have less text and fewer keywords in the headlines and titles, yet still rank well.
Advertisements: Too many advertisements, especially AdSense ads, can have a negative impact on search engine results pages.
As a non-statistician I feel compelled to pass along the Search Metrics caution about drawing definitive conclusions about correlation equaling causation from this data. The above mentioned factors may or may not be being used by Google as SERP signals. They acknowledge that a “question like ‘does a site receive social signals because it ranks well or does it rank well because it receives social signals?’ are absolutely valid and cannot be answered unequivocally with the current data.”
What are your experiences regarding social influencing search results?
Thanks for visiting. As always comments and questions are welcomed and appreciated!