Yesterday I published a well received article on econsultancy on why Google can be fairly confident that Google+ could already give them good social signals for rankings, even though they don’t have as many users as the other social media sharing sites.
I won’t rehash that article, but my gut feeling was, G could be fairly confident if something was popular Google+, it would likely to be popular on other social sites. But gut-feeling is rarely enough. So I carried out tiny scale correlation study looking at one site and whether there was any mathematical relationship between the number of shares on Google+ and the other social sites.
Initially I took 15 recent articles from SearchEngineLand and benchmarked the number of Tweets, Google+1, Linkedin Shares and Facebook Likes. The results were startling.
I explored whether there was a linear correlation between the number of Google+1 votes and the other social votes. There was a correlation a really, really strong one. There were dozens of flaws in this sample, it was just one site, just fifteen articles and the audience of SearchEngineLand far more likely to use Google+ than any ‘normal’ web users but there was no argument the correlation was there.
Between G+ and FB 0.97
Between G+ and Tweets 0.94
Between G+ and LinkedIn Shares 0.95
1.0 is a perfect correlation, and this kind of correlation is pretty much amazing, so much so initially I had to check my sums several times. At this point I’d made my point for the initial blog post but so amazed by the results I had to explore further.
While something being a great piece of content should increase the likelihood of being shared, i.e. all good stuff gets shared but surely different pieces of content would appeal more or less to certain social media users.
So I expanded the test to include a couple of other sites. Surely that’d be enough to break the correlation.
This time I added a further two popular blogs, this time Mashable and & Engadget still fifteen articles. (They didn’t include LinkedIn stats so I dropped studying them) and the correlation was still really strong.
Correlation between Tweets & FB 0.66
Correlation between Tweets & G+ 0.81
Correlation between FB & G+ 0.69
So not quite as strong as the initial sample but still pretty strong, I considering expanding the study to more sites and more articles but I’m growing in confidence there is a huge correlation between the two.
Would you find the research useful?