Page 1 of 4 123 ... Last
  • Jump to page:
    #1
  1. Contributing User
    SEO Chat Explorer (0 - 99 posts)

    Join Date
    May 2014
    Location
    Somewhere between your modem and your browser
    Posts
    48
    Rep Power
    23

    SEO Lessons: Lesson 1, how to take apart a search engine


    I've read through a lot of threads on this forum and I thought I would write a quick post to clarify some points about SEO and search engines as I see them. While you might not agree with them all, I've been around the block since 1999, and so this might be helpful as a general kind of overview, with some advice thrown in for people that are coming at the SEO thing afresh.

    How to rank in search engines?
    Back in the good 'ole days it really was easy to rank in the search engines. Before Google there was Bing, Altavista and Looksmart directory. Bing or MSN actually delivered very good results with very little efforts. It was the same with Google when they finally arrived. What was consistent back then was the actual amount of time you would spend researching how the search algo. worked. I can remember putting up many different domains with lots of differently constructed templates, some with a meta-description of X length, some with a meta-description of Y length. Sometimes with frames loading inside an external JavaScript, sometimes text on the page being hidden from the view of browsers, sometimes masking landing pages as search results and sometimes replace.history for breaking making it a better user experience. The goal was always the same, bring people to your website or that of your paying client. In the forums people might suggest what was working, but everyone knew that no-one was about to give up their secret recipe on how they were ranking websites. Today people have magic formulas that come with a fee. Some of the pay-per-lead affiliate networks can offer upwards of $40 or $60 for getting people to apply for mortgages or loans, so ask yourself this question when you are next reading a Google traffic solution - If this person could easily bank this kind of money, why are they not doing this by getting amazing results in Google instead of just making me spend $20 on a tool to take the first page?

    Researching the search engines and not telling anyone
    One of SEO industries biggest failures is its inability to have taught new people coming through how to break apart a search engine, and to actually work out how the search engines are ranking their pages. I have talked to many many SEOs that are simply clueless. They come out with questions and you find yourself sitting their thinking "oh my dog, we have to go THAT FAR BACK, just to get them on the page". That's because it is the exceptional rather than the rule, that your budding SEO will start with a learned blog rather than a blank piece of paper when starting the whole de-fragmenting of a keyword space. The SEO industry seems to have failed so badly that we see the whole concept of SEO being morphed into something else which covers broader bases like on-site conversion measurement, and performance. Don't get me wrong, I'm not pissed at anyone, I'm saddened that we've allowed this situation to happen, so all those people from back in the days we should be ashamed of ourselves.

    I'm therefore going to share some really elementary steps which I hope some of the readers can start with and then take their results from there.


    Lesson 1 - Analyzing how a search engine ranks pages.
    There are lots of factors that impact on how a search engines ranks pages. Depending on the level of your skills and interaction with SEO, the metrics that get churned out by industry are things like Google Page Rank, Domain Authority, MozRank, Links, Alexa ranking, Trust rank, blah blah blah di blah blah etc! As mentioned when I was testing search engines back in the day we had the luxury of instruments that meant withing 48 hours we could check to see how the pages were ranking. This meant that in the space of a week we usually had a search engine algo down completely. Unfortunately we don't have that situation today but we can still do some tests if we have some time, and frankly if you're doing SEO for long term you need to make sure you are doing R&D.

    Here is an example of how to measure the impact that on-page content will have on rankings. We are going to make the following statement: When all other things are equal, what is the best keyword density to make pages rank on Google.

    Step 1: Pick a keyword.
    First we are going to pick a keyword. We are not interested where this keyword ranks, but we are interested in being able to guarantee that irrespective of the page optimization, that we can find it in the Google or Bing index against the targeted keyword. Probably one of the best things is to therefore invent a keyword that is unique and bring back very few results in Google.

    Step 2: Register 5 domain names
    We are not going to test keywords in domain names, so we register 5 domain names that bear absolutely no relevance to the keyword we are targeting. Go to the waybackmachine and verify that the domains have not been used in the past and then whois.sc/domain.com your domain to make sure it has never been registered. I might register 195039213556.com 986983815.com for example.

    Step 3: Get the content.
    Depending on your writing skills you may be able to produce this content in-house or you might need to farm it out. Pick a theme for your content and then produce 5 pieces of text that you will simply paste into the main text area of your webpage. The most important thing is to make sure that your content has different weightings of keyword density in the page area. David Nylor has a tool that you can use for this purposes. Once you have written the page past the text into a web-page and do not apply any formatting save for a paragraph that encapsulates the text block and upload it to the domain.

    If you are using the tool I've cited here, enter your keyword and the URL of the website and then click "see all" underneath the keyword analysis on the results page. All you should be looking at on the subsequent page is the TEXT column percentages. What you want to do is make sure that you have a range of different keyword densities. Your exact method you can define yourself.

    You might say: What is the keyword density required for a page that has 400 words. Or you might say, what is the keyword density for a page that has 200 words, you can define that. To make sure that the page ranks, you should put the keyword in the TITLE of the page. Just the keyword. That's the great thing about testing, you are in control. Set your own measurements!

    Step 4: Recapping
    To recap on where we are at, you should:

    1. Have registered 5 domains and have Webspace for them. They should all be on the same web-host.
    2. Have created 5 pages of content with different keyword weightings and have those as the one-page websites.
    3. If you have a copy of Webceo, you can setup a ranking report where all your domains are setup in a project as competitors and you can then run ranking on the domains against the keywords on a daily basis against the keyword to monitor rankings.

    Step 5: Indexing
    The best thing to do in terms of getting the pages indexed is just to leave them, Google or Bing will find them eventually. If you try to speed up the indexing then you need to make absolutely sure that you apply exactly the same level of linking to every domain you own. That means if you want to add the links to the domains from another already indexed domain you can do this, but you need to put all the domains there.

    Results
    Eventually (depending on how much you decide to shoe-horn the indexing process) you are going to know the keyword density of the page that gets the highest ranking. You now know what the ideal relationship of text is. In terms of costing this process, it's going to cost you a maximum of $60 to get this answer. It might take 3 months for the pages to get indexed, but if you want to be the best in your industry this is what you need to do. Once you've done that test case, you can perform similar controlled experiments on the domains, simply rolling out different optimization elements on the pages that are already ranked. Just make sure that you are controlling exactly the variables.

    That's how to take apart a search engine.

    LittleG.
  2. #2
  3. Digital Marketing
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Aug 2008
    Location
    Florida
    Posts
    8,044
    Rep Power
    5750
    Google doesn't use KW density as an element to determine where to rank a website for a given query
  4. #3
  5. Contributing User
    SEO Chat Explorer (0 - 99 posts)

    Join Date
    May 2014
    Location
    Somewhere between your modem and your browser
    Posts
    48
    Rep Power
    23
    Originally Posted by KernelPanic
    Google doesn't use KW density as an element to determine where to rank a website for a given query
    Okay, I'm glad you think that but unfortunately it falls somewhere in the void that I was talking about in our industry.

    I'm not of that opinion because a) I test this test periodically get results on this and b) once you strip away all the external factors people say impact on search engine rankings you are always going to be left with page that a search engine parses and will be scored based on a number of factors. No-one is saying that this is a magic formula to page 1 of Google or Bing, it's about taking one element of SEO (on-page) and showing here how to test it.

    Unfortunately we live in a read and believe kind of world rather than a truly analytical scientific one, that's what's required for SEO and that's what we've lost (although I have no way of testing this as we talk in a forum).

    If you audit the rubbish that has been written on this topic you immediately get people talking about "spammy keyword stuffing". Filter the authority sources in these niches against the products they are seeking your subscriptions for and see what's left over, I can tell you it's not much. What's the fallout of this? People thinking that putting keywords in their pages is going to make them rank well. When everyone is doing the same thing no-one stands out and that's great for people like me and others that do proper testing.

    KW stuffing is exactly what this is not about, but it's far easier to write this stuff discredit SEO as simply a bunch of KW stuffers when in fact this is just the easy-lets-demonize the industry approach benefiting all but a few.

    There's a really great video I saw where a very famous person starts by talking about frequency of words as an indicator of determining the relevance for a page, and then says that anything after that is kind of spammy. Now considering that it is the same company that generates nearly all it's revenue from it's paid advertising platform, the question must be asked:

    "As a rational, intelligent human being can I exclude the fact that, considering the multi-stakeholder position of its business, that the information being provided to webmasters could conceivably be slanted towards an interest that is not principally in favour of letting a webmaster achieve a top-3 place in the unpaid search results?"

    How much assumption is based on someone else's conjecture rather than a qualified test?

    Lots.

    LittleG

    Comments on this post

    • GabrielG agrees : Good argument
  6. #4
  7. rod@missionop.com
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Mar 2003
    Location
    Palm Beach Gardens FL 33410
    Posts
    16,979
    Rep Power
    0
    Originally Posted by LittleG
    Okay, I'm glad you think that but unfortunately it falls somewhere in the void that I was talking about in our industry.
    Get a grip!

    You get points for keyword in the title element, in the page header, in the body text, even in the domain name, and in the URL, and other parts of the path... and then you add all that up and compare all that to stuff that has absolutely nothing with ranking for "this phrase" ... and you believe you get more points for that?

    Use a little common sense!
  8. #5
  9. Contributing User
    SEO Chat Explorer (0 - 99 posts)

    Join Date
    May 2014
    Location
    Somewhere between your modem and your browser
    Posts
    48
    Rep Power
    23
    Originally Posted by fathom
    Get a grip!

    You get points for keyword in the title element, in the page header, in the body text, even in the domain name, and in the URL, and other parts of the path... and then you add all that up and compare all that to stuff that has absolutely nothing with ranking for "this phrase" ... and you believe you get more points for that?

    Use a little common sense!
    You're so right, I don't know what I was thinking. Please ignore this post of mine clearly a moment of madness.

    Can't wait to read the scientific "point" approach to SEO. Will it also include advice with phrases like "frowned upon"

  10. #6
  11. No Profile Picture
    Contributing User
    SEO Chat Discoverer (100 - 499 posts)

    Join Date
    Aug 2012
    Location
    Huddersfield, UK
    Posts
    164
    Rep Power
    164
    Into the lions den I go. ** sidles in **

    I do not believe that there is a magic percentage for keyword density that will result in you ranking well. Full stop. ** looks around, no lion yet **

    I do believe that keyword density plays a part in SEO however ** was that a shadow, gulp **

    When we say don't keyword stuff, it will hurt your ranking and perhaps get you penalised by a fluffy animal, what are we talking about? Keyword density perhaps? ** Was that a growl behind me **

    In the end, in my opinion, is quite simple - keyword density is NOT a quick way to rank. There is NO MAGIC FORMULA (which is where problems with forums etc generally come in) that you can apply and say that 'now you will do well'. Unfortunately most people talking about wanting to know what 'the keyword density should be' are still hoping that there is a way they can get top rankings with no effort and no cost (which, if there was such a formula, everyone would be using anyway so we'd be back to square 1) and are the same people who expect to buy 1000 links for $5 and get #1 spot.

    However, there is something to be said for ensuring you don't stand out from the crowd.
    If, at one end of the scale, you have TOO high (whatever that may be for your keywords/niche/etc) a keyword density (stuffing) then you are almost certainly going to either: not do as well as you would otherwise have done, or, possibly, even tank completely.
    If you have, at the other end of the scale, no instances what so ever of your keywords or synonyms then you are almost certainly going to not do as well as you could be. (Please note this is just 'likely not do as well' and is not saying what change that would be. Good links, well structured website, and 100s more ranking signals are still in play!)

    So, as with most things, unless you are really going to spend some serious time and effort on testing etc (in the hope that there really is a formula and you can find it) then the best advice for almost all website owners who look for 'keyword density answers' is, as it always has been really: write good content, that reads well, about the topic in question, using the keywords and synonyms naturally in titles and copy and forget about keyword density as an issue completely.

    (Please note: I'm a numbers freak and love testing theories out :- so it could be a case of 'do as I say and not as I do' ;) )
  12. #7
  13. Contributing User
    SEO Chat Explorer (0 - 99 posts)

    Join Date
    May 2014
    Location
    Somewhere between your modem and your browser
    Posts
    48
    Rep Power
    23
    roooooaaaaaaarrrrrr!!!! (he escapes unscathed).

    Correct. No magic formula.
    Correct. KW density is not an easy way to rank. Different percentages different niches.

    Anyone else wanna come in the pit
  14. #8
  15. Digital Marketing
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Aug 2008
    Location
    Florida
    Posts
    8,044
    Rep Power
    5750
    It's hard to imagine that it's 2014 and we're arguing about extensive testing to determine the best kw density. It makes me sad. In 2008 it was a fun debate, now it's just pathetic.
  16. #9
  17. Super Moderator
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Mar 2004
    Location
    Gloucester (South West UK).
    Posts
    6,533
    Rep Power
    3522
    Hi LittleG,

    Originally Posted by LittleG
    As mentioned when I was testing search engines back in the day we had the luxury of instruments that meant withing 48 hours we could check to see how the pages were ranking. This meant that in the space of a week we usually had a search engine algo down completely.
    2 things wrong with this:
    1. "Back in the day" search engines didn't update their results every 48 hours, they did it periodically; you had to wait for up to 3 months to see any change in results.
    2. If you (ever) "had a search engine algo down completely" you must be a) Famous (in the SEO world). b) Incredibly Wealthy.

    Your advice on "how to take apart a search engine" is to test each element of the algorithm. I won't argue with that, it's pretty obvious that testing is a good idea (that's why we have a test & experimentation forum). Sadly you cannot test every element of any SE algorithm, some of the more advanced (and meaningful) elements (like back link profile and domain history) become beyond your control once they have been on line for any length of time, since you cannot stop others from linking to the site or citing the domain.

    However, picking on "keyword density" as your first example was a mistake imo, plus, you haven't actually cited any real test results, you've simply advised people to test, and inferred that there is such a thing as an "ideal keyword density".
    I too have been working in, studying and testing SEO since "back in the day" and things have changed beyond all recognition... (kw density used to matter a lot).
    Since then things have moved on greatly, to the point where - apart from the obvious - on-page factors pale into insignificance in the overall scheme of things, and keyword density is amongst the least important of all on-page factors.

    If you're going to reverse engineer a search engine algorithm, start with things that really matter... and if you're going to infer that keyword density makes a difference, give some statistics and real data, not just hearsay, inference and argument.
    Last edited by ClickyB; Jun 10th, 2014 at 09:01 AM.
    ClickyB
    "The quality of the visitor is more important than the volume..." (Egol 22nd Feb 2008)
    [New to SEO/SeoChat?] [Canonical Problems?] [Forum Rules & Posting Guidelines]
  18. #10
  19. SEO Insultant
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Oct 2005
    Location
    sharing a room with my ego
    Posts
    6,470
    Rep Power
    3421
    I'm not convinced yet that Google does NOT take kw density into account

    {/advocate of the devil}

    Just sayin'....
  20. #11
  21. Digital Marketing
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Aug 2008
    Location
    Florida
    Posts
    8,044
    Rep Power
    5750
    while your busy hooking your horse up to your buggy all those things whizzing past you at 80 MPH are cars. You should think about getting one
  22. #12
  23. SEO Insultant
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Oct 2005
    Location
    sharing a room with my ego
    Posts
    6,470
    Rep Power
    3421
    Ok I'll step it up a notch: I DO think that Google considers kw density in SOME way....

    P.S.: but please also note all my previous posts about kw density and the pursuing of the perfect percentage thereof.... :-)
    Last edited by dzine; Jun 10th, 2014 at 04:27 PM.
  24. #13
  25. Digital Marketing
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Aug 2008
    Location
    Florida
    Posts
    8,044
    Rep Power
    5750
    OK Mr step-it-up-a-notch, seeing as how you are the new Advocate General for keyword density, how important an element of SEO do you think it is? Where does it fall in with title elements, header tags, internal link anchors, url, image alt text, keyword meta tags (lol)... Where do you put it in order? And while we're at it, what exactly is the perfect KW density?
    Last edited by KernelPanic; Jun 10th, 2014 at 05:22 PM.
  26. #14
  27. Digital Marketing
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Aug 2008
    Location
    Florida
    Posts
    8,044
    Rep Power
    5750
    Originally Posted by LittleG
    it falls somewhere in the void that I was talking about in our industry.
    The only void in our industry in the one between people's ears who want to call themselves experts
  28. #15
  29. SEO Since 97
    SEO Chat Mastermind (5000+ posts)

    Join Date
    Mar 2011
    Location
    Arizona
    Posts
    8,764
    Rep Power
    5665
    Is it keyword density or merely content density?
    IMHO the term keyword density should be changed to content density, which in my mind is a lot more important.
Page 1 of 4 123 ... Last
  • Jump to page:

Similar Threads

  1. Will Search Engine Marketing - Replace - Search Engine Optimization
    By -search-engines-web in forum Search Engine Optimization
    Replies: 9
    Last Post: Feb 27th, 2013, 01:53 AM
  2. Replies: 4
    Last Post: Feb 4th, 2012, 01:49 AM
  3. Replies: 0
    Last Post: Feb 18th, 2009, 08:00 AM
  4. MSN Search: Learning Lessons
    By SEO Chat in forum SEO Chat Articles
    Replies: 0
    Last Post: Apr 13th, 2005, 08:00 AM

IMN logo majestic logo threadwatch logo seochat tools logo