The increasing influence of social signals on search engine rankings was the key takeaway from Searchmetrics 2012 survey on UK SEO Ranking Factors.

The survey looked at 10,000 selected top-keywords, 300,000 websites and millions of links, shares and tweets, and compared potential ranking factors with their Google rankings.

The key findings

It’s important to note that a correlation does not necessarily represent causation, but the summary points of the report are worth reading:

  • Social media signals have arrived in the UK: social signals from Facebook, Twitter and Google+ now correlate very strongly with good rankings in Google’s index.
  • Too much advertising is detrimental: just as new in the UK is the observation that pages with too many advertisements have a harder time achieving high rankings. However, the problem apparently relates only to AdSense adblocks.
  • Backlinks are still important but quantity is not the only important thing: even though the number of backlinks is still the most powerful factor, links with stop words and ‘nofollow’ should also be included in the link-mix.
  • Brands leverage classic SEO signals: apparently pages with strong brands do not need be as concerned with the areas of title tags, headings etc.
  • Keyword domains still frequently attract top results: despite all the rumours to the contrary, keyword domains are still alive and well and are often in the top rankings.

Google+ is also a factor

It’s also worth noting that there was a very high correlation with Google +1s, but this was discounted due to the low level of activity.   However, they note that Google is trying to make Google+ an important player and watch out for further developments.

More links correlate with higher rankings

One very clear conclusion is that while the quality of links is becoming more important, sites with a high number of links are likely to rank more highly.  This graph shows this clearly.

Number of inbound links correlates with rankings

What about Penguin?

It’s important to note that this research was carried out in February and March 2012 and does not refer to the April Penguin update.  It would be interesting to see the effect on this graph.  Our guess is that the level of correlation would be significantly reduced.
By Iain Martin