1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Criteria For Analyzing Keyword Competition

Discussion in 'White Hat SEO' started by Diophantus, Sep 2, 2014.

  1. Diophantus

    Diophantus Junior Member

    Joined:
    Aug 20, 2014
    Messages:
    130
    Likes Received:
    22
    Occupation:
    Housekeeping
    Location:
    Texas
    Through my research, I can't really find a focused answer to this, so I wanted to get some opinions:

    What indicators should one look at in sizing up their competition in the google top ten rankings of their choice keyword?

    A couple that I have been looking at recently were the obvious ones, such as the number of backlinks. I imagine that if there is a site with tens of thousands of backlinks, that ranking might be difficult. However, if this were the case, then the top ten would likely be placed in the order of the the number of backlinks, which certainly appears to not be the case. So there are some other indicators.

    Off the top of my head, I'm looking at:

    - Content
    - Backlinks
    - Keyword density in articles?
    - various rankings such as Alexa, Moz, PR, Page Authority, etc.
    - Social signals (likes, pluses, tweets etc)

    Of course if I am too specific on this question, my real question becomes "what's Google's Algorithm" which I believe everybody would like to know the answer to.

    If anyone has some insight on the power or relevancy of some of these indicators, I would appreciate the input.
     
    • Thanks Thanks x 1
  2. ardley216

    ardley216 Elite Member

    Joined:
    Mar 28, 2008
    Messages:
    2,391
    Likes Received:
    2,356
    Occupation:
    Finding easy keywords
    Location:
    1,500,000,000 Keywords Re
    Home Page:
    Hi.

    This is an interesting question... I've spent the last year or so FOCUSED on.. "what's Google's Algorithm".. and I have been trying to figure this out through keyword research. In fact I did my thesis on this exact question for my university in London.

    Anyway, I did large scale research comparing up to 40 different metrics from individual metrics like exact match results to software generated metrics like LTP's difficulty rating.

    One thing I found is that social signals has the highest correlation to top ranks... but I also found that this is more of an effect of being highly ranked rather than the cause. If you think about it logically.. the more important the page is.. the more likely google will rank it at the top.. So what is 'importance?'

    Well, obviously for a page to be important it has to be important to a particular keyword, lets say.. "blue widgets". This particular page needs all the right ingedients to be #1.. And these can be:

    relevant content
    useful content
    built on relevant website
    other people need to find it useful
    It needs to be recognized as useful by other authority sources
    People need to feel the need share it with their peers

    Now, to measure this, we can use the metrics as you mentioned above, plus others..

    such as:

    what anchor texts are used.
    linking page authority
    bounce rate
    linking domain authority
    keyword in title/url/description
    etc
    etc

    So what I did was get more than 10,000 keywords and compare ALL these metrics against each other, I analysed the top 10 for each keyword and the 90-100 keywords to compare with the top 10.. and see what made the difference.

    Essentially.. I found that More backlinks does not equal better ranks.. its purely down to 2 things:

    What backlinks you have and how good your website is.

    I know this seems pretty obvious but its important to understand that if you have one backlink from a very reputable source, this will give you MUCH MORE authority than a million crappy GSA backlinks.

    And before you get links.. You need to build a website that people will WANT to link to.. The content has to be unique, interesting, consistent, useful etc etc... Once you have this as a foundation you are ready to go.

    I know this is kinda obvious, but my research proved this IS the case. Good websites do better than bad websites.. DUH!! One of my use case's was MatthewWoodward.com and how he now ranks for "SEO Tools" without building a single backlink. Useful websites win!!

    I hope that helps you in understanding "what's Google's Algorithm".. That said... All I did was open a pandoras box... I'm addicted to figuring out Google's Algo through the eyes of keyword research... :)
     
    • Thanks Thanks x 14
  3. ardley216

    ardley216 Elite Member

    Joined:
    Mar 28, 2008
    Messages:
    2,391
    Likes Received:
    2,356
    Occupation:
    Finding easy keywords
    Location:
    1,500,000,000 Keywords Re
    Home Page:
    Sorry I went on a tangent..
     
  4. Diophantus

    Diophantus Junior Member

    Joined:
    Aug 20, 2014
    Messages:
    130
    Likes Received:
    22
    Occupation:
    Housekeeping
    Location:
    Texas
    Actually Ardley, I think your tangent was exactly where I needed to be. I was also considering looking at a large number of indicators, not necessarily to learn Googles algorithm but to do two things:

    1.) See how many of these indicators follow a Pareto Distribution when compared against each other.
    2.) If they generally do, perform large scale research. Use the data points (say between some competitition indicator and keyword length) and derive a set of equations that can be used as a general measure to statistically determine whether or not to use a particular keyword.

    My general assumptions are that the use of keywords and SEO are similar to exploring statistically the natural laws of the universe, ecosystems, etc. Therefore, some general laws could be derived on what kind of keywords can be used.

    Its possible that through the use of statistical analysis, googles algorithm could be reverse engineered or at least well understood and exploited.

    Oh and I'm interested in your keyword service too. Can you give a sample keyword for my specific niche? I want to see how good your keywords are :)
     
    Last edited: Sep 2, 2014
  5. Diophantus

    Diophantus Junior Member

    Joined:
    Aug 20, 2014
    Messages:
    130
    Likes Received:
    22
    Occupation:
    Housekeeping
    Location:
    Texas
    BTW, I completely understand your addiction. Ever since I started getting into IM again, I have been OBSESSED with keyword research. Its a natural high to see a page rank. Same thing with finding that "sweet spot" keyword with low comp and high volume.
     
  6. IMpossible

    IMpossible Supreme Member

    Joined:
    Apr 15, 2012
    Messages:
    1,338
    Likes Received:
    302
    Occupation:
    Internet Marketing Guru
    Location:
    Somewhere on earth
    Very well said!
    Regarding the link profile I'd say that 5 High Quality links beat 500 crappy blog comments anyday.
    But you always need to consider that you need to have a natural link profile, and this would include blog comments aswell.
    There's no mega popular website on this planet which has no crappy blog comments.
    Why? Because you simply can't decide what the users (who obviously enjoy your site) do with your URL.
    They don't come to you and ask: "Hey, can I mention your website on this blog's comment section?"
    They just do it. And this is how natural link profiles start to evolve. That's it.
     
  7. ardley216

    ardley216 Elite Member

    Joined:
    Mar 28, 2008
    Messages:
    2,391
    Likes Received:
    2,356
    Occupation:
    Finding easy keywords
    Location:
    1,500,000,000 Keywords Re
    Home Page:
    Yeah, you've got the right idea, or the same ideas as I had when I began.. The real struggle I had was getting this data accurately on a large scale, but once this issue has been cracked, it gets a little easier to get large enough samples to test data.

    One intersting finding I found that software generated difficulty metrics were often WAY off - such as LTP and UNF. I need to do more testing on TT.

    One side thing I found interesting when doing keyword research is observing search trends. For example, my business partner is expanding his web design services to weddings, but found no one wanted wedding websites in June, when all the weddings were taking place.. I did some research and found that the highest searches were in January for buying keywords for weddings. This meant he could prepare over the winter for a full launch.

    Yes I believe that statistical analysis will give at least half the story of google's algorithm.. but I do know that actually ranking for any given keyword is the only way to determine Google's algorithm.. Of course, assuming you can isolate each variable, which is almost impossible.

    Sure, I'll send you a sample right now. I don't have niche specific samples though, but do check out my sales thread in my Sig.



    oh yeah I know what you mean. I just research all day every day, when you find those 100,000 search keywords that you can rank with $10 of fiverr links its epic!! :)

    Oh yes you are definitely right there. When it comes to long term link building you have to completely imitate other authority sites in your niche. Some backlinks WILL be low quality, and have anchor texts completely irrelevant to your niche, these are important as well. I do find that good relevant blog comments still rank very nicely for my websites. But the main aim is to get your visitors to build you links for you! MAke the website and content so useful they share the page on peoples blogs, in blog posts, wikipedia etc etc.

    Hope that helps! :)
     
  8. Diophantus

    Diophantus Junior Member

    Joined:
    Aug 20, 2014
    Messages:
    130
    Likes Received:
    22
    Occupation:
    Housekeeping
    Location:
    Texas
    What's this? I'm interested :)
     
  9. ardley216

    ardley216 Elite Member

    Joined:
    Mar 28, 2008
    Messages:
    2,391
    Likes Received:
    2,356
    Occupation:
    Finding easy keywords
    Location:
    1,500,000,000 Keywords Re
    Home Page:
    Hey I send you a PM :)
     
  10. stackcash

    stackcash Registered Member

    Joined:
    May 19, 2011
    Messages:
    50
    Likes Received:
    11
    Occupation:
    WordAgents
    Location:
    Long Island, NY
    Here's what I look at when deciding if I should tackle a new keyword:


    • Domain authority of the competition
    • Page authority of the page I'm trying to out rank
    • Age of competitors domain
    • Number & quality of inbound links pointing to competitor websites
    • Whether or not a competitor is ranking with a private blog network, or some other asset that gives them an advantage
    • Number of direct competitors in the niche
    • The amount of onsite optimization that was done within the top 10 results
    • Potential search volume vs potential profit
    • Number of "big brands," big box stores, and e-commerce sites in the niche that may have a competitive advantage

    If all looks good after checking out those metrics, I'll usually create a small site in the niche for an easy-to-rank longtail version of the main keyword. If I can frontpage the longtail easily, I'll go ahead and start working on the main/money keyword.
     
    • Thanks Thanks x 1
  11. ardley216

    ardley216 Elite Member

    Joined:
    Mar 28, 2008
    Messages:
    2,391
    Likes Received:
    2,356
    Occupation:
    Finding easy keywords
    Location:
    1,500,000,000 Keywords Re
    Home Page:
    Thats a good list stackcash,

    One thing I would say is that high domain authority isn;t always a bad thing. Reason being is that when there is NO competition for a given keyword, that does NOT mean there are no results in the top 10.. The top 10 results will then be filled with websites with high DA such as amazon, Yahoo Answers and so on.. These pages will have no PA, and therefore no backlinks, they are likely to not be optimised for that keyword. This means it is EASY to out rank.

    On the other hand, if you find the top 10 results have domains with 10-30 DA and 10-30 PA.. this means that LOADS of websites are completing for this keyword. Every difficulty metric I have come across has missed this out.

    So with regards to DA and PA... its not a one size fits all situation.
     
  12. stackcash

    stackcash Registered Member

    Joined:
    May 19, 2011
    Messages:
    50
    Likes Received:
    11
    Occupation:
    WordAgents
    Location:
    Long Island, NY
    Yeah, it's definitely an art rather than a science. Always a balancing act.

    Do you have a process for finding those High DA/No PA keywords?

     
  13. ardley216

    ardley216 Elite Member

    Joined:
    Mar 28, 2008
    Messages:
    2,391
    Likes Received:
    2,356
    Occupation:
    Finding easy keywords
    Location:
    1,500,000,000 Keywords Re
    Home Page:
    Yeah it can feel like an art, but I always believe EVERYTHING is a science, you just got to figure it out! ;)

    Right now.. No. I am looking for a way though. If anyone has any ideas, please message me!!
     
  14. mootonandy

    mootonandy Supreme Member

    Joined:
    Jun 21, 2010
    Messages:
    1,257
    Likes Received:
    236
    Just go with number of USA searches per month, as the searches per month goes up then so does the competition generally.
     
  15. ardley216

    ardley216 Elite Member

    Joined:
    Mar 28, 2008
    Messages:
    2,391
    Likes Received:
    2,356
    Occupation:
    Finding easy keywords
    Location:
    1,500,000,000 Keywords Re
    Home Page:
    Yes there is a trend like that. But this IS not the case whatsoever. I am targeting a keyword with over 350,000 searches and not made the backlinks yet, and I am on page 4 - brand new site. If you do correct keyword research you WILL find keywords that will flood you with cash.
     
    • Thanks Thanks x 1
  16. mootonandy

    mootonandy Supreme Member

    Joined:
    Jun 21, 2010
    Messages:
    1,257
    Likes Received:
    236
    Brand new domain?
     
  17. natewex

    natewex Power Member

    Joined:
    Sep 22, 2009
    Messages:
    602
    Likes Received:
    127
    Location:
    Ireland
    Dan, please PM me in regard to this also...