29 Jun Google SEO Ranking Factors 2013
Searchmetrics.com recently released its 2013 Google ranking factors (PDF) study. The study sifts through a large volume of search metrics to find factors that influence a website/webpage’s ranking in the search engine results pages (SERPs).
I feel it is necessary to provide some sort of disclaimer. This study does contain statistics taken from the real world, but stats can be misinterpreted by readers, including me. With that, I found the Internal Linking section (the PDF not the infographic) of the study a little confusing, particularly the description of “internal links” and “self-referential” internal links. I’m not suggesting that there is something wrong with the data presented, but the way the findings are described was confusing to me. Maybe another cup of coffee will help.
There are some great, easily understood, takeaways in the ranking study (PDF) that can be incorporated into your own SEO efforts, but because I am interpreting someone’s interpretation, please draw your own conclusions and ask your own questions. As the nerdy kids say: “RTFM”.
Search data used in the study
SearchMetrics used 10,000 search terms to generate data for the study excluding informational queries for the most part. The 10,000 searches resulted in the following data:
- 30,000 SERPs
- 300,000 titles, descriptions, URLs, etc.
- 150 GB of data
- 600,000 AdSense blocks
- 5,300,000,000 backlinks
- 4,150,000,000 Facebook shares
- 12,950,000,000 Facebook likes
- 600,000,000 Facebook comments
- 1,000,000,000 tweets
- 330,000,000 Google plus ones
- 14,500,000 Pins
Source: SearchMetrics SEO Ranking Factors – Rank Correlation 2013
Overview of 2013 ranking factors
The study findings described the relative ranking factors found in four main areas:
- Social signals
- Technical on-page factors
The influence of social signals continues to grow and have a major impact on ranking. The following factors have a strong influence on ranking:
- Number of Google Plus Ones
- Facebook Likes
- Presence and activity on social networks (sharing)
The number of backlinks to a page remains a highly influential factor but natural backlinks profiles outperform “cooked” profiles.
- Quantity of backlinks
- Higher percentage of no-follow and backlinks with stopwords than exact match anchors
- Quality of linking page (SEO visibility of linking page)
- Markedly reduced exact match anchors
Technical on-page factors
On-page technical factors that correlate to SERPs ranking are:
- Keyword presence in the title tag
- Keyword position in the title tag Leftmost performs better.
- Presence of keyword in h1
- Presence of keyword in description
- Mere presence of h1, h2, meta description The mere presence of h1, h2 and description show a positive influence on ranking regardless of whether they contain keywords.
- Page load speed (marginally positive)
- Keywords in domain/URL have little influence on ranking.
- Google authorship integration had virtually no influence on ranking though SearchMetrics suggests that it may become more influential in certain realms where authorship is an important part of discerning the web page’s relative value (Newspapers? Authoritative Blogs? Subject experts?)
- Keyword relevance of a page
According to the study, the power of keywords in the the domain name or URL is not relevant to to ranking.
…the quality of the keyword in the URL/domain as a Ranking Factor is no longer relevant SearchMetrics Ranking Factors 2013
Though I’d like to believe this is the case (as it levels the playing field), I still see many keyword domains at the top of SERPs (search engine results pages). Appearances are deceiving, though. It may be that other factors are making these exact-match domains rank so well.
On-page content continues to have less power than social media or backlinks but good on-page optimization is still considered a basic prerequisite of page ranking. The content ranking factors in order of influence are:
- Number of internal links Pages with a higher internal link count outrank those with a lower link count. The study showed that on average, a number 1 ranking web page contained between 90 and 100 internal links (huh?). Let the confusion begin (for me). I’m assuming for clues in the study that “internal links” refers to any links including those found in site-wide navigation.
The average number of links will naturally increase in the case of websites with strong menu structures and many internal links in the footer SearchMetrics 2013 Ranking Report
Something about the numbers in the SearchMetrics links/rankings graph above hurts my head. A few years ago, Matt Cutts stated that a reasonable number of links was fewer than 100. This became the standard widely regard by all who follow Cutts’ gospel. I should say that “we” made it the gospel. Mr Cutts did not appoint himself prophet of the SEO throng. That widely quoted comment was updated by Mr Cutts based on a rapidly evolving Web. The “100” was also removed from Webmaster Guidelines which now reads: “Keep the links on a given page to a reasonable number”. What’s reasonable? I supposed if the number of links prevents a good user experience, you have too many.
- Pages ranked #1 had an average HTML character count of approximately 60,000. The curve rises steeply to an average of 80,000 characters for pages ranked #3 then begins its gradual fall. The rise between 1st and 3rd position are influenced by what SearchMetrics calls the “brand factor”. Websites with a very strong brand can outperform those with less brand presence.
- Word count
The average #1 ranked page was 476 words in length. This seems to be the sweet spot. As the study states, one would expect a continuously falling curve from this position, but again the “brand factor” influences the shape of the curve. Pages with extremely strong brand can rank with a lower word count so the word count rises to an average of 599 words for website’s ranked #3 and then begins its gradual fall.
- Keywords in body
- Keywords in external links Avoid using the keyword for which you are trying to rank as the link anchor text in links pointing to an external site.
- Self-referential link anchors Web pages with #1 ranking had, on average, one link to themselves. Web pages with #3 ranking had on average one self-referential link. Now, this is not a major on-page factor it seems, but again, I find that that the description a little confusing (it’s me not the study).
From the pool of websites that have at least one internal link, seven out of ten URLs ranked in the 1st place have one link to themselves on average, while websites that rank in the 3rd place contain on average one self-referential link. SearchMetrics Ranking Factors 2013
The graph in the study then illustrates a decline in ranking when the “self-referential” link count is marginally below 1. A website with just below .6 self-referential links was ranked around 29th.
- Keyword in anchors show a positive correlation to ranking.
The current rule of thumb suggests to use “hard” links internally (= with keyword/s) and “soft” links externally (= generic / stop words / more words in the anchor, brand links, etc.) SearchMetrics Ranking Factors 2013
This is inline with what Matt Cutts had to say about exact match anchors on-page.
- Image count The presence of embedded images and images influence rank. The more images, the better. The study states that this is not limited to images embedded in content but images in general (including images in the website’s design).
- Keyword in h2 A relatively small correlation between ranking and the presence of a keyword in an h2 heading seems to exist.