Infographic: 2013 SEO Ranking Factors, From SearchMetrics

The folks at SearchMetrics have released their SEO ranking factors for 2013. This is their ranking correlation study they worked on for the past 3-months, right after the second-generation Penguin update was pushed out by Google. Want to understand everything that was in it? They’ve got an infographic for that (click to enlarge):

us_ranking_factors_2013

The key takeaways from their study included:

  • Keyword domains and keyword links have lost relevance
  • Brands are the exception to many rules
  • Social signals continue to correlate very well with better rankings
  • Good content is always important: it comes down to quality
  • The number of backlinks remains immensely important
  • On-page technology remains one of the basics

There’s been some criticism around this study on social networks, but like with any ranking correlation study, there are always critiques, and there should be. You can download the study over here.

Also be sure to see our Periodic Table Of SEO Success Factors, which was freshly updated earlier this month, for our own take on ranking factors to consider.

Related Topics: Channel: SEO | Google: SEO | Infographics | SEM Industry: Stats | SEO: General

Sponsored


About The Author: posts interesting infographics related to search engines and search marketing from around the web. Opinions and facts shared by these infographics are those of the original authors and not by Search Engine Land. Want your infographic featured? See this page.

Connect with the author via: Email



SearchCap:

Get all the top search stories emailed daily!  

Share

Other ways to share:
 

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. Comments may also be removed if they are posted from anonymous accounts. You can read more about our comments policy here.
  • Chase Anderson

    I’ve never seen a less accurate rank factor report. Keyword in Title has 0 relevency alone is absurd. Social shares simply don’t impact 99% of organic rankings. Unless you’re dealing with a hot news topic, it’s still all about #1 links, #2 relevance.

  • http://www.seo-theory.com/ Michael Martinez

    Since they don’t know which links pass value in the search indexes their conclusions are highly dubious.

  • http://kercommunications.com/ Nick Ker

    I think that Keyword in Title = 0 must be a mistake in the chart. I skimmed the whitepaper to find details about that, and it looks like the chart just doesn’t match what they are saying and what we all are pretty darned sure about – that words in titles matter.
    I have seen more sites/pages doing well without exact keywords in the title, but nowhere near enough to say that keywords in the title have significantly less importance.

    That said, there does seem to be other sketchy info here. They do point out that correlation does not equal causation, but it is at the bottom of the infographic and somewhere in the whitepaper. Since people who read and parrot SEO news rarely check the details before declaring something the new mega-important must-do thing, I think we can expect a lot of “Keywords In Titles No Longer a Ranking Factor” posts to appear in 3… 2… 1…

  • RyanMJones

    Pet peeve, must mini rant:

    Likes, comments, plusses, etc aren’t a ranking factor. matt said so. It’s more likely that sites that rank high get more traffic because of this ranking. more traffic means more likes, plusses, comments, etc. We’re looking at the correlation backwards.

    Also.. let’s talk correlation. To be statistically significant you need a correlation coefficient of .8 – by definition. .4 is a “week correlation” – and every one of these factors is under .4

  • http://www.keshkesh.com/ Takeshi Young

    Agreed. There may be a strong correlation between social signals and rankings, but not causation. Listing it as a top “ranking factor” is misleading.

    Also, keywords in the URL are still a huge ranking factor. Despite the EMD update, exact match domains continue to rank very well.

  • RyanMJones

    I’d like to see what their correlation coefficient is for “having a title tag” and “HTML tag on site” I bet they’re also in the .3 to .4 range.

  • Rob Wagner

    I agree with you Michael. When you don’t know the value the conclusions are more like guesses. In their defense the Infographic is pretty.

  • RyanMJones

    it’s because of how they look at data. we know that it helps, and because of that a higher percentage of spam sites actually do it compared to to non-spam sites which may not have keywords in H1 even though it could help them.

  • Chase Anderson

    I was so inspired, I decided to cover more details with the problems with this report. http://www.clicksandclients.com/2013-rank-correlation-report/

  • Chase Anderson

    Yet another inherent flaw in the report.

  • Chase Anderson

    Nope – they would be 0 or less, since it’s a consistent variable. If 50% of the pages had and HTML tag and 50% didn’t it’d be in the .3 to .4 range :).

  • http://www.seo-theory.com/ Michael Martinez

    There have been several “correlation studies” recently from different sources, all making completely unsupportable claims about the effects (or lack of influence) of links on the SERPs. These unscientific claims continue to feed the SEO community’s addiction to nonsense analysis. One would have hoped that people would have learned better by now.

  • Marcus Tober

    @NickKer:disqus The infographic displays the correlations. If everyone includes the keyword in the title tag, because it is a basic SEO thing, then the correlation is neutral, so 0. The study shows what have good ranking sites in common and differs them from worse ranking sites.

  • Marcus Tober

    I want to point out, that some of you only looked at the graphics and didn’t read the text around. :) The study shows what have good ranking sites in common. Low correlations could mean, that it is a low or no ranking factor it also also mean that every site is having this signal. Then the correlation is near 0.

    I can highly recommend to read the whitepaper, because all the factors and the metrics behind are well explained. We refreshed the data for the after Penguin 2.0. So you can see the impact of Penguin, because we implemented in the whitepaper a comparison before and after Penguin. Again, it’s worth to read it.

  • Marcus Tober

    @Michael_Martinez:disqus Nobody knows which link passes value. But with enough data you get a good picture what have good ranking sites in common. You don’t need to know then which link passes value…

  • http://www.seo-theory.com/ Michael Martinez

    You absolutely need to know which links are passing value because if you cannot determine that then your analysis is based completely on random guesses and therefore has no credibility.

  • emarketing .

    2013 is about Content Marketing. Quite an impressive report. Thanks for sharing!

  • Marcus Tober

    Michael, sorry but you are wrong. Search engines determine rankings on many variables not just external links. If a site ranks many different variables were taken into account. Our study shows what good rankings have in common. One of the results is that good rankings have on an average more words in their text. Better rankings are using the keyword that was queried in title-tag, body etc. This has nothing to do which not knowing if an external link passes value or not. We analyzed 10.000 informational search keywords. We took into account billions of external links, also billions of internal links, billions of social signals etc. It has nothing to do with guesses Michael. It is the result of a study made by our data scientists. I can highly recommend reading the whitepaper.

  • Marcus Tober

    Will, the keyword in the H1 tag has a negative correlation. That means that less sites with good rankings have the keyword in H1, but many sites from position 3 down to 30 have the keyword in H1. So the conclusion is that mainly brand sites that don’t have the best onpage optimization rank really good because of the brand bonus (Panda etc). It doesn’t mean that having the keyword in H1 is negative. It only has a negative correlation. I can highly recommend reading the whitepaper. You will see for H1 the raw correlation and then it is clear why it so low.

  • http://kercommunications.com/ Nick Ker

    Ah, got it. Like I had mentioned previously, the whitepaper made sense to me, but the infographic had me and others thinking you were nuts :)

    That’s the thing about infographics, I guess – even people like me who do read the details, will look at pretty pictures and see a low bar on a graph under a big “Ranking Factors” title and think “0 = unimportant”.

    Takeaways here:
    1) Distilling a huge amount of info into a graphic that makes it all clear to people who will look at it for just a minute is not as easy as it may seem.
    2) Never underestimate the SEO community’s ability to take one bit of information, misinterpret it, and then crow about it.

    I have read the whitepaper and highly recommend it to anyone interested. Thanks @Marcus Tober and SearchMetrics.

  • Michael Thomson

    If I +1 a page and my friend see’s it ranking higher with my annotation, is that not a ranking factor. I suspect your reference was to non-personalized search, a slowly eroding away facet of search. One must make that distinction.

  • http://www.seo-theory.com/ Michael Martinez

    Until you can show which links are passing value, your conclusions about how much links are having an impact are invalid. All the wishful thinking and rationalization in the world will not change that.

  • http://www.keshkesh.com/ Takeshi Young

    It’s still pretty easy to get EMDs to rank with zero backlinks for their keyword terms for low-medium competition keywords.

  • http://www.keshkesh.com/ Takeshi Young

    Yeah, the title of this infographic is completely misleading. These are not “ranking factors”, it’s just a bunch of correlations.

  • Marcus Tober

    Sorry no, we don’t show how much links having an impact. This is not the case for the chart and also not for the whitepaper. We show what good rankings sites have in common aka correlations. It absolutely doesn’t matter for the correlations which link passes value.

    And Michael, a website can also rank without even having an external link. Links are only a part of the whole equation. If there are no external links, there many other signals like relevance of the document, onpage signals etc.

  • Nikhil Raj. R

    This is great infographic only if one reads the PDF report of the study! If you see the Onpage factors, the Keyword in title has 0 but position of keyword in title has 0.12 correlation. Thats strange! But thinking deeply ;) and reading that PDF I realized that they were saying almost all the top pages in search results have keywords in title. So thats not a positive but 0 correlation. But the position of keywords have a high correlation, that means pages with keywords to extreme left of the title tag does get higher rankings. So the title should contain keyword and it should be to the left of the tag.

  • http://www.seo-theory.com/ Michael Martinez

    Then you should retract the “ranking factors” claim for the whole study.

  • http://www.adbot.in/ Adbot India

    Keyword domains has been lost ranking is the impact factor as also in the recent study and live experiment by Adbot, we found that the keyword domains will jump to a considerable ranking on the launch but looses away within few days.

    A clear and good study to revise our SEO strategies. Thanks!!

  • http://www.stanleyoppenheimer.com searchengineman

    Michael Martinez seems to lay down the Gauntlet – (Not as angry as Michael Grey!) :-D

    It would be nice to see a rebuttal from Mr. Marcus Tober. on Michaels Blog, regarding his points shooting down studies like this. The argument is over my head. But I am savy enough to know that Correlation is certainly not Causation. “Lambrogini’s are red, therefore fast cars are red!” I am realistic. That being outside Google’s Black Box.

    Here is a fun Avinash graph which shows this in action.
    https://plus.google.com/u/0/+avinash/posts/XdBQK2mrDMX
    (PS Facebook certainly does not match Myspace!)

    Perhaps a (Respected) third party who both parties can agree on, can weigh in on this. Many strategies and future plans are swayed by powerful info-graphics like this.

    You got my attention…

    Far be it for me to believe everything I read online. Is the argument about the methodology of the test? Or the way the info-graphic is presented (Misleading?). Can the results of this study be tested in some way? or is this impossible due to the complexity of Google’s Algo (Waste of time?)

    Blog: http://www.seo-theory.com/2013/06/27/madness-reigns-in-the-state-of-confusion-in-online-marketing/

  • alchemyv

    Agreed with Michael Martinez. Furthermore, any statistician worth their salt would consider that anything is positively correlated if the squared error (R squared coeff) is over 0.66. Anything below that implies it doesn’t correlate and is just noise. The highest correlation here is 0.34, which implies that attempting to correlate with rankings is likely to fail.

  • Mark Goho

    I’m not sure why, but there’s almost no point in publishing correlations that are less than 0.35….to say that there is a correlation of 0.16 or 0.2 is basically to say that there is no correlation at all…it’s just so low! I get absolutely nothing from this report.

  • Ajay Rana

    Hi Marcus,
    Can you please send it to me at ajayrana0001@gmail.com. I would be really thankful to you.

  • alchemyv

    If I may, I strongly suggest readers check the whitepaper at MathSight.org as it produced a study specific to Penguin 2.0 and the findings are based on 90% statistical confidence using variance analysis.

Get Our News, Everywhere!

Daily Email:

Follow Search Engine Land on Twitter @sengineland Like Search Engine Land on Facebook Follow Search Engine Land on Google+ Get the Search Engine Land Feed Connect with Search Engine Land on LinkedIn Check out our Tumblr! See us on Pinterest

 
 

Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States

Europe

Australia & China

Learn more about: SMX | MarTech


Free Daily Search News Recap!

SearchCap is a once-per-day newsletter update - sign up below and get the news delivered to you!

 


 

Search Engine Land Periodic Table of SEO Success Factors

Get Your Copy
Read The Full SEO Guide