searchmetrics email facebook github gplus instagram linkedin phone rss twitter whatsapp youtube arrow-right chevron-up chevron-down chevron-left chevron-right clock close menu search
60336033

The 2014 Rank Correlation Analysis and SEO Ranking Factors for Google U.S.

It’s here! This fall season’s most important piece of SEO reading. This year’s report, available as a PDF to download, took a little longer than usual to prepare, because it’s more than a study. It’s an extensive analysis of trends and the status quo in SEO today, and we are publishing it as a White Paper. It runs to 100 pages, has easy-to-follow chapter headings, and covers what we think are the most important factors in SEO 2014.

RF - Header

We have taken on board all the feedback and critiques that we received from previous studies, and incorporated these into the White Paper. It’s now more than a White Paper, with a scope and an academic approach that has produced the 2014 fall season’s SEO survival guide for SEOs and online marketers in the field.

 

White Paper Download:
Rank Correlations And SEO Ranking Factors Google U.S. 2014

 

We have greatly expanded the number of features researched to analyze and evaluate what differentiates the higher-ranking sites in the top positions in Google from those that don’t quite make it.

This year we have put more weight on average values of the features, as these produce a more accurate analysis in addition to the respective correlation values. In doing so, we want to prevent potential causal inferences, the tempting but often incorrect ‘gut feel’ analysis that we at Searchmetrics avoid at all costs.

That’s why we focus on these two approaches:

  • Correlation of Top 30 = Differences between URLs within SERP 1 to 3
  • Averages of Top 10 = “Similarities” of top sites ranking on SERP 1

The Right Interpretation: Correlations, Averages And Ranking Factors

As an extra this year, to be as transparent as possible in explaining our approach Correlation Versus Causationand the background to this study, we have a separate section explaining exactly how we define the term “Ranking Factor”: What is it? How are our correlations calculated? What conclusions can we draw from these calculations, and what are the actual average figures we analyzed for the features we investigated? All of this can be found here: What is a Ranking Factor?

It’s well worth looking at the examples that illustrate our workings, the explanation of the data we used and our analytical approach – the backbone of our findings.

Features, Results And Evaluations for 2014

Searchmetrics Ranking Factors - SERP 1to3We have analyzed the top 30 Google rankings for 10,000 keywords – approximately 300,000 URLs – with a focus on information-oriented search terms. Correlations relate to the measurable differences between individual search results, and this year we have included these in the bar chart below, which shows correlation values for the most interesting features of 2014:

Searchmetrics Rank Correlations 2014

We have, of course, examined factors from previous studies, but this year we have included additional analysis of many new features, particularly in the areas of Content, Backlinks and Brands.

New Features: Content, Backlink And Brand Sections

Some of the new features are intrinsic to the Content Optimization area of the Searchmetrics Suite, and are based on co-occurrence evaluations of keywords and relevant terms. For the first time we have included user features such as Bounce Rate measured on a keyword basis, which provides information to correlate user signals with rankings. This data came from thousands of Google Webmaster Tools accounts and was anonymized for the analysis.

White Paper Download:
Rank Correlations And SEO Ranking Factors Google U.S. 2014

I highly recommend downloading and reading our white paper, the “SEO Must-Read 2014”.

By the way, we’ve even had a factor dealing with “http vs. https” within this year’s pool of features to be analyzed, but we have decided not to take it into the study. Even after updating the analysis there seems to be no relation between the search results and https evident (yet?). I have already written something about it here: HTTPS vs. HTTP – Analysis: Do secure sites really get higher rankings?. We will keep you updated.

 

2014 – Main Developments And Trends At A Glance:

For the impatient among you, here are the notable findings of the 2014 White Paper, including rank-correlation analyses, averages and Ranking Factors in a nutshell:

1 Content

  • There is a measurable correlation between the quality of content and rankings. This is demonstrated by, among others, the analysis of two new features based primarily on word co-occurrence analysis: Proof and Relevant Terms.
  • The length of content continues to increase.
  • A good internal linking structure is an important factor, and probably the most underrated SEO measure.

2 Onpage Technical SEO

  • Onpage, the keyword remains an important part of the overall concept for SEO, often represented by a balanced presence in Title, Description, Body copy, H1, H2, etc. Needless to say, that keyword stuffing should still be avoided. However, there is a definite trend towards developing keywords to topics to generate holistic content.
  • Site load speed is a veryRF Robot important performance factor.
  • Good site architecture is the beginning and end of effective SEO.

3 Backlinks

  • The quantity and especially the quality of backlinks remains important.
  • The number of keyword backlinks continues to decrease, even if the correlation increases.
  • Backlink features for Brands (Point 6 below) appear to work differently to the rest of the URLs in SERPs.

4 Social Signals

  • There were minor changes to the previous year, with Social signals correlating slightly less with good rankings than last year.
  • Average values rose slightly.

5 NEW* User Signals

  • Both the click-through rate and the time-on-site are considerably higher in better ranking sites – this may appear obvious, but average values determined over many URLs can be used as a benchmark for your own optimization.
  • The bounce rate is lower for top-ranking URLs.

6 Improved* Brand Factor

  • There seem to be special consideration for big brands.
  • The Brand Factor and its definition have been revised this year, to show the increasing complexity of its influence and quality.

Conclusion: SEO In 2014 And Beyond – What Are The Takeaways?

There are areas where Google seems to have found a good balance. However, there are many more areas where constant revision and development of features is taking place, with the treatment of Brands constantly maturing. The real focus, however, is still page content, where relevance is key. In all areas, the trend towards a natural structure is clear. Just as the quality of the link profile is better when designed, the overall content structure is better when it is easily understood and holistic.

Here’s the download link one more time:

White Paper Download:
Rank Correlations And SEO Ranking Factors Google U.S. 2014

SEO is becoming more and more natural, with the focus moving towards Search Experience Optimization (= the user) in the future, and with the technical aspects of site optimization remaining crucial. Let us have your thoughts on this year’s evaluation? As always, I look forward to your feedback and suggestions!

Marcus Tober

Marcus Tober

My name is Marcus Tober and I’m the founder of Searchmetrics. Because we really love to analyze all kinds of online data, we can give you more insights than any other company in SEO, SEM and Social Media. It’s not a job, it’s passion.

36 thoughts on “The 2014 Rank Correlation Analysis and SEO Ranking Factors for Google U.S.


  • Marcus,

    You said the number of keyword backlinks continues to decrease I agree with you but the point is How can we up our site on page # 1 with a specific keyword? I have read tons of articles but never understand this thing that without giving weight to the keyword how can we go up in the word?

    I hope you will answer me the best.

    Thanks

  • Thank you for a very insightful and analytic study. I also like the comments about social signals and corresponding links to searchenginewatch.com articles. A good example of a great content that gets shared 🙂

  • Référenceur Grenoble 2015/02/02 at 2:11 pm

    Thank you for this relevant article comprising for the essential and objective analyses on the factors of ranking current!

  • This is a great case study, has already benefited greatly ..
    It will help people to understand the factors that search engines
    Thank Marcus

  • Watching as several programmers are developing CTR bots. Should be interesting to see how this plays out.

  • I may be off the mark, off my head or perhaps need to head off to a new career, but – and I have revisited this blog post a few times recently to confirm Facebook shares and Facebook comments are as high up the list as they are- for everyone of my SEO clients, Facebook signals plays ZERO part in their ranking factors. In fact, I have yet to see sufficient evidence that amazing social media campaigns plays any part in significant SEO changes. I have no doubt that social media might play a part in newsworthy SEO rankings (such as celebrity, major local, national or international news items, in terms of Google reflecting what people are buzzing about by putting sources to the top of rankings) but for small businesses trying to get up the rankings, I see no major impact. The way I see it, time spent on social media is time not spent on SEO. Am I wrong or way of the mark?

  • Hi James, don’t mistake correlation for causation. The bar chart shows the calculated values of the spearman correlation for these factors. We created a page with more explanations on how to interpret the data here: http://www.searchmetrics.com/what-is-a-ranking-factor/

  • In practice I still see a fair bit of manufactured links and small private link networks in the small business eco system and it still seems to be working better than it should but the long term direction is clear – relevance, supporting content, marketing, PR & user experience should be at the top of your to-focus-on list.

  • Hi,

    I have read through your ranking factors and I am very surprised that you have given Google +1s more weight than backlinks when listing rank correlations from top to bottom.

    Can you explain further?

    Regards

    Brian

  • Hi Brian,

    the bar chart is not ordered by weight but by correlation. This is a huge difference (see also my comment above). We took a different approach of presenting the correlations as an infographic this year. And we also added a “relevance score”. Check it out here: http://www.searchmetrics.com/knowledge-base/ranking-factors-infographic-2015/

  • Do you have the checklist for 2015 ?

    Thanks in advange

  • I am not yet familiar with this and I have read through Your ranking factors and I am very surprised that you have given Google + 1s heavier than backlinks when listing rank correlation from top to bottom.
    And do you have a checklist for 2015?

  • very very great information and one of the most useful case studies
    thanks marcus

  • actually, I really liked your article with because it is very creative and very interesting, because I was so amazed with the results of your work and your blog is very useful for everyone

  • I am not yet familiar with this and I have read through Your ranking factors and I am very surprised that you have given Google + 1s heavier than backlinks when listing rank correlation from top to bottom.
    And do you have a checklist for 2015?

  • thank you for information In practice I still see a fair bit of manufactured links and small private link networks in the small business eco system and it still seems to be working better than it should but the long term direction is clear

  • حلويات جزائرية 2016/01/25 at 8:19 pm

    thank you for the great ifos

  • thanks for ths great free informations

  • The data collect from the correlation of Top 30 = Differences between URLs within SERP 1 to 3 and averages of Top 10 = “Similarities” of top sites ranking on SERP 1,is indicative but not in alf for explain the real correlations.And Google+ is not a signal anymore.

  • Great article although its a couple of years old it still holds up well in 2017.

  • Agree that it still holds up well in 2017 but is there an updated seo ranking factor list for 2017? I’ve seen other big seo company research for 2017 but would be good to also see your in-depth version.

    Thank you.


Write a Comment

Note: If you enter something other than a name here (such as a keyword), or if your entry seems to have been made for commercial or advertising purposes, we reserve the right to delete or edit your comment. So please only post genuine comments here!

Also, please note that, with the submission of your comment, you allow your data to be stored by blog.searchmetrics.com/us/. To enable comments to be reviewed and to prevent abuse, this website stores the name, email address, comment text, and the IP address and timestamp of your comment. The comments can be deleted at any time. Detailed information can be found in our privacy statement.