SitePoint Sponsor

User Tag List

Results 1 to 3 of 3
  1. #1
    SitePoint Evangelist worksdev's Avatar
    Join Date
    Mar 2002
    Location
    Central, PA - originally from Monterey, CA
    Posts
    497
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Domains and URLs-Rankings

    http://www.sitepronews.com/archives/2002/june/10.html

    I was just reading the above article and want to know what you all think. Are there any potential holes in the findings?

    Should this make us re-think our URLs and domains?

    I wonder how much of it has to do with the human factor. Google page rank popularity takes into account the number of other pages linked to a page. I wonder if a page URL that is too long to type is less likely to be linked to.

    What are your thoughts?

    Best Regards,
    worksdev

  2. #2
    Serial Publisher silver trophy aspen's Avatar
    Join Date
    Aug 1999
    Location
    East Lansing, MI USA
    Posts
    12,937
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Thats the most bogus study I've ever seen.

    Just because kids who have many appliances at home tend to do better in school does not mean that appliances make you smart.

    First of all you have to remember that search engines and directories are different things, people too often combine them when they make their claims and thats wrong.

    Does having the keyword (ie: search term) in the domain name
    affect ranking?

    Yes:- This was the factor that was found most consistently among the top ranking sites. Since this single factor was found so consistently among the top ranking sites, we strongly recommend that you choose a domain name which contains your major keyword. Our own domain doesn't follow this rule, but you can bet that we are considering moving to a domain which does include our keywords now that we have reviewed the findings of this study.

    Sure it helps, but why does it help? Does it help with search engines because its an extra keyword, or because when people link to you via your URL (<a href = "site.com">site.com</a>), which is a very popular way to link to a page, you've got a keyword in your anchor text.

    Does having a URL shorter than average affect ranking?

    We found consistently that shorter URLs ranked better. This is an interesting factor since you can largely control the length of your URLs without moving to a new domain. Were you already considering a restructure/redesign of your current site? Consider using short directory names and filenames.

    Your domain name is also a factor in the length of your URL. At one time, many search engine optimization experts were recommending purchasing very long domain names with lots of keywords separated by dashes. The results of this study indicate that would be bad advice. Instead, focus on short domain names which contain your main keyword and short URLs.
    Again, this person can't separate the cause and effect of the situation. Is the long site ranking bad because it is long, or because less people are linking to it? Or perhaps is the short site ranked high because it's owned by a large corporation who pours millions into their site? Most short generic domain names are worth alot of money.

    And again, differentiate between search engines and directories. The hyphens are very important because directories, and maybe even some search engines, cannot parse keywords contained in runon words like webmasterresources. Yahoo would parse "webmaster" but not "Resources" DMOZ would parse neither.

    You don't get a keyword rich domain for search engines, you get it so you can rank high in Yahoo.

    Does having a "/" at the end of your URL affect ranking?

    This is one of the most interesting findings. We actually began by studying lots of page types (.html, .htm, .asp, .shtml, .sht, etc.). Our data kept showing that we were missing a large percentage of the top ranking pages by just looking at those file types. We were puzzled at first. Everyone knows that .html is the most common ending to URLs; right? Wrong!

    Actually, we found that most top ranking pages end with a reference to a directory, not an actual .html file. For instance, we found the following:

    http://www.domain.com/test/ - This was the structure we found most often among the top ranked sites.

    http://www.domain.com/test.html - This structure was more likely to be found among the lower ranked sites.

    We think this is a key find. If you are considering a restructure of your site, you might want to consider renaming test.html to index.html and placing it in a subdirectory called "test".
    This study was obviously done by someone without a scientific testing background because again they attribute an incorrect cause to their effect. Are top level directory pages more ranked higher by default, or are they usually promoted more?

    For instance which page do you think Sitepoint promotes more?

    http://www.sitepoint.com/

    or

    http://www.sitepoint.com/privacy.php

    Also which one do you think is more apt to have incoming links? And finally do you honestly think this has anything to do with the "/" at the end of the url.

    We expected many top-level domains to possibly rank lower than the standard ".com" domains. What we found surprised us. Many domain endings that we expected would rank lower seemed instead to be more-or-less equally distributed among the high and low ranking sites.

    We never expected the ".net" domains to be ranked lower, but that is what the data suggests. In fact it is a very strong correlation. We used to think that a ".net" domain was almost as good as a ".com" domain. We will now avoid them.
    Again, is this because of the domain name or because the people with money to promote buy a .com? This is like saying that because a Grand Am is the most popular car in America it is also the "best."

    This is an interesting find. We used to have many sites with this extension. Everyone is familiar with ".html" files; right? You may have also seen ".shtml" files. The ".shtml" extension is often used to denote a file that should have Server Side Includes (known as SSI) processed prior to serving the page. It used to be common to use the ".sht" extension interchangably with the ".shtml" extension in much the same way as ".htm" is used interchangably with ".html"

    Warning! In our 20,000+ survey of sites listed by the top search engine, we found zero sites with a ".sht" extension. This seems to show that the top search engine does not list ".sht" files at all!
    Another example of their brilliant logic. I've been making websites for 7 years and I've never heard of anyone using .sht, I've been on the Internet for 8 years and I've never seen .sht. Search engines must also ban .pancake extensions too because I bet they didn't find any in their list.

    Does having a ".html" URL affect ranking?

    This is really just the opposite finding as one of the findings among sites that rank higher. Since URLs ending in a "/" are found so often among the top ranking sites, it turns out that URLs ending in the standard ".html" are among the lower ranking.

    In fact, this file extension seemed to be even more associated with lower ranking sites than ".shtml", ".htm" and ".asp". Prior to this report, we considered the ".html" at the end of a URL to be the most common and desirable. THis study has changed our mind about this as well.
    More brilliant deductions. Chances are the professionals with money and resources and better sites used more advanced technology than straight html, so they will be using other extensions on their site.

    I believe they teach logical reasoning in highschool geometry. You have exercises such as:

    John likes beef.
    Dogs like beef.
    Therefor John must be a dog.

    Whoever wrote this article/study must have been gone that day.
    Chris Beasley - I publish content and ecommerce sites.
    Featured Article: Free Comprehensive SEO Guide
    My Guide to Building a Successful Website
    My Blog|My Webmaster Forums

  3. #3
    SitePoint Evangelist worksdev's Avatar
    Join Date
    Mar 2002
    Location
    Central, PA - originally from Monterey, CA
    Posts
    497
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Thanks Aspen:

    I was hoping you'd show up and give comments.

    I agree with everything you said about separating the cause and effect.

    I would also add:

    There is no attempt to define how they arrived at the definition or criteria for top-ranking sites, what key word categories and industries they were top ranking in, and on what search engines (or as you pointed out) directories they have top ranking on.

    Best Regards,
    worksdev


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •