8 {{ upvoteCount | shortNum }}
8 {{ upvoteCount | shortNum }}

Page Content, Algorithms and Query Relevance

by raiko
7 replies
  • SEO
    • |
A couple of months ago GodOverYou wrote a reply to a post and stated:

"Actually with the advent of Hummingbird and how Google now breaks down a page's content into portions (which is how they algorithmically determine theme and relevancy to a query) it's much easier to rank an 'autoblog' with truly 'unique' content."

Can someone give me a quick idea of how Google breaks down page content to determine query relevancy and theme? I assume he means something beyond <h> tags and footers and headers?

Beyond that, how is knowing this "portioning" of pages useful to someone building webpages?
#algorithms #content #page #query #relevance
  • Profile picture of the author yukon
    Banned
    Hummingbird didn't really change much, example, Wikipedia was breaking pages down into sub-sections years ago. Look at how Wikipedia usually gets jump-links added to the SERP description, that's partially how pages are broken down in order to target multiple search queries per page.

    Hummingbird was mostly ask/answer search queries, Googles version of Siri (Apple).
    {{ DiscussionBoard.errors[9417653].message }}
    • Profile picture of the author raiko
      Does that imply that pages are potentially ranked better when structured in a manner similar to Wiki's? Is that because Google's algorithm has an easier time assigning relevancy when it's spoon fed to them through a "Contents" menu such as Wiki uses?

      I mean Wiki utilizes that content menu on long articles and structures everything with section numbers such as:

      1.0
      1.1
      1.2
      1.2.1
      2.0

      But, their actual article text doesn't even employ the section numbers.
      {{ DiscussionBoard.errors[9417775].message }}
      • Profile picture of the author yukon
        Banned
        Originally Posted by raiko View Post

        Does that imply that pages are potentially ranked better when structured in a manner similar to Wiki's? Is that because Google's algorithm has an easier time assigning relevancy when it's spoon fed to them through a "Contents" menu such as Wiki uses?

        I mean Wiki utilizes that content menu on long articles and structures everything with section numbers such as:

        1.0
        1.1
        1.2
        1.2.1
        2.0

        But, their actual article text doesn't even employ the section numbers.
        Structured pages do have advantages (ex: SERP jump links, Google Sitelinks, Google Mega Sitelinks, multiple ranked internal pages per keyword, etc...), but the real advantage is when you link multiple structured internal pages that are all focused on the same subject.

        Those numbers you see in the Wikipedia TOC (Table Of Contents) are seperated from the actual keyword in the Wikipedia HTML source code, example (source):

        <li class="toclevel-1 tocsection-15"><a href="#Alternatives_to_the_automobile"><span class="tocnumber">13</span> <span class="toctext">Alternatives to the automobile</span></a></li>
        You can simplify your own pages by omitting those TOC numbers, they're not necessary. The TOC itself is very useful for both traffic & SEO.
        {{ DiscussionBoard.errors[9417831].message }}
        • Profile picture of the author raiko
          Yeah, it's interesting how Wiki does it. I looked their Barack Obama page and noticed they use the TOC with a jump link to the section Early Life and Career. Click it and, of course, you jump down to the section Early Life and Career which, right below that, has a link to a separate wiki page titled Early Life and career of Barack Obama. Both pages have similar content.

          Do you actually utilize Wiki's internal linking structure? Sometimes their articles are difficult to read given all the hyperlinks embedded in the text. It reminds me a little of the ebay affiliate Skimlinks if you're familiar with them. They hyperlink anything in your page text that is available on ebay and make it an affiliate link. I would guess Wiki has an automated way of creating those hyperlinks as well.
          {{ DiscussionBoard.errors[9417879].message }}
          • Profile picture of the author paulgl
            Wikipedia is forced to because they have only one page to get everything on it.
            Without targets, the page would be an unbearable mess to google and humans.
            (they do break some things up, like discography for music artists)

            I, on the other hand, face no such pressure and break long pages into smaller bites.

            Paul
            Signature

            If you were disappointed in your results today, lower your standards tomorrow.

            {{ DiscussionBoard.errors[9417894].message }}
            • Profile picture of the author raiko
              The longer Wiki articles are definitely borderline unbearable to me. I've never really been able to decide whether it's their page structure that contributes the most to their search engine rank or the fact that everybody and their dog links to their articles as if they are the end all for truthful and accurate information.

              In the past I, personally, tended to build my sites with the smaller "bites" you are referring to. Each page would cover a specific piece of information. However, I have recently written some page content that I purposely left very long and wrote specifically for the visitor such that all relevant information regarding the subject would be included on that page. Those pages have generated more traffic more quickly than other pages I've written. I assumed it was because the page ranked for multiple terms though I haven't checked. A brief page wouldn't have that ability.

              Wiki has an advantage in that they cover every subject known to humankind. Most other websites focus on a particular subject, so, the internal linking that can be done intelligently is somewhat limited. Nonetheless, I might try adding a higher number of links from internal page to internal page just to see what effect it has.
              {{ DiscussionBoard.errors[9417941].message }}
  • Profile picture of the author yukon
    Banned
    You don't have to have long pages to be able to use a TOC.

    You could have something as simple as an FAQ page with 5 questions/answers (example), where the TOC jumps down the page to the matching content.

    Another example of a simple page is, building an image gallery on one page where a TOC jumps down the page to the matching image, that gives you an optimized page with no articles.
    • TOC jump link includes keyword anchor-text
    • Keyword in <h2> tag above image (title of image)
    • Image includes keyword alt-text
    • Image has keyword caption text below image
    • Image caption text has keyword internal link pointing to a relevant internal page
    • Repeat with the next image/keyword...

    When I suggest to use Wikipedia as a guideline I'm talking about page structure, the Wikipedia content is irrelevant.
    {{ DiscussionBoard.errors[9419383].message }}

Trending Topics