Social Signals Again Seen Highest Correlating Factors With Google Rankings

Searchmetrics-Google-US-Ranking-Factors-June2013Within the US, among 44 ranking factors examined, social signals account for 7 of the 8 most highly correlated with Google search results, according to a new study from Searchmetrics. Presenting the correlations using Spearman’s rank correlation coefficient, the study determined that Google +1s (0.4) have easily the highest correlation, followed by number of backlinks, Facebook shares, and Facebook total (summary of shares, likes and comments), each at 0.34. Last year’s study, which looked at 21 factors, found social signals occupying 5 of the top 6 spots.

That social signals correlate with Google rankings is not news to search marketers. According to a recent survey from BrightEdge, 80% of search marketers believe social sharing of content will be either a much more (31%) or more (49%) important means to improve rank this year than last.

In the Searchmetrics study, other social signals correlating with Google rankings included Facebook comments (0.33), Facebook likes (0.31), Pinterest pins (0.29) and Tweets (0.28). (Google +1s and Pinterest pins were new to the study this year.) Still, the report cautions that social signals – as with other factors – cannot be said to be causal, but only correlated (i.e. well-ranked domains have many shares, +1s, etc.). Even so, the researchers cite studies suggesting that indexing is possible only through social signals.

Aside from social signals, backlink-related factors also continue to show up in the study as having a positive correlation with rankings. The number of backlinks was one of the highest-ranked factors, and other related factors found to have positive correlation included the proportion of nofollow links (0.25) and the proportion of backlinks with a stop word (0.18). The study notes that when combined with other factors, backlink factors tend to produce a very positive correlation.

Other Findings:

  • On-page coding factors appear to be a basic prerequisite for a high ranking.
  • Unlike in last year’s study, this year keyword domains have decreased in importance and do not seem to be relevant for ranking. However, Google does not seem to have the same standards for brands as non-brands, such that there are some domains for which the factor “keyword in domain” has a more positive correlation. In fact, overall, there seems to be a running theme that Google strongly prefers brands and ranks them highly even if they do not meet some criteria.
  • Keywords in the description, title and H1 are still important, and the researchers recommend placing them close to the front of the title.
  • While Searchmetrics expected author integration to have a strongly positive correlation with rankings, that was not the case. The researchers still believe it will become more relevant.
  • Word count in text has become a more positively correlated factor since last year, though the number of internal links (0.15) is the most highly correlated on-page content factor.
  • Positive correlation for the number of images on the website (more = better) has increased from last year.
  • Whereas integration of ads (AdSense and other Adlinks) resulted in a negative correlation last year, this year the correlations are virtually neutral.

About the Data: The report cautions that correlation does not equate to causation, and these results do not in any way guarantee that the factors have an effect on rankings or are even used by Google as signals.

Searchmetrics largely excluded specific navigational keywords (as was the case last year) from the original data pool, which included the top 10,000 keywords according to search volume, not wanting to distort the analysis. Search queries are considered to be navigational keywords if they return results, which are largely irrelevant and only return one exact match (for example: “Facebook Login”).

The first three pages of organic search results were always used as data pool. That keyword set covers approximately 90-95% of the data pool used in the study last year. Here, the researchers attempted to strike a balance between two factors, namely the preservation of the “greatest common denominator” as an optimal basis for comparison with last year’s analysis, and the consideration of new keywords that have emerged due to growth in search volume in the top 10,000.

The researchers note that their data pool is always kept up-to-date. Therefore, new, relevant keywords are used in current studies, such as, for example “Samsung Galaxy S4″ or “iPhone 5,” which did not even exist at the time of the 2012 study.

The 10,000 analyzed keywords led to:

  • 30,000 SERPs with 300,000 titles, descriptions and URLs; and
  • Approximately: 150 GB in data; 600,000 AdSense blocks; 5.3 billion backlinks, 4.15 billion Facebook shares, 12.95 billion Facebook likes, 600 million Facebook comments, 1 billion tweets, 330 million Google plus ones, and 14.5 million pins.