There are hundreds of factors that help engines decide how to rank a page. And in general, those hundreds of factors can be broken into two categories—relevance and popularity (or "authority"). For the purposes of this demonstration you will need to completely ignore relevancy for a second. (Kind of like the search engine Ask.com.) Further, within the category of popularity, there are two primary types—domain popularity and page popularity. Modern search engines rank pages by a combination of these two kinds of popularity metrics. These metrics are measurements of link profiles. To rank number one for a given query you need to have the highest amount of total popularity on the Internet. (Again, bear with me as we ignore relevancy for this section.)
This is very clear if you start looking for patterns in search result pages. Have you ever noticed that popular domains like Wikipedia.org tend to rank for everything? This is because they have an enormous amount of domain popularity. But what about those competitors who outrank me for a specific term with a practically unknown domain? This happens when they have an excess of page popularity. See Figure 1-1.
Figure 1 -1: Graph showing different combinations of relevancy and popularity metrics that can be used to achieve high rankings
Link Popularity
en.wikipedia.org/ get,adobe,com/reader/ awesome.com/
I Domain Popularity ED Page Popularity
Althoughen.wikipedia.org has a lot of domain popularity and get.adobe.com/reader/ has a lot of page popularity, www.awesome.com ranks higher because it has a higher total amount of popularity. This fact and relevancy metrics (discussed later in this chapter) are the essence of Search Engine Optimization. (Shoot! I unveiled it in the first chapter, now what am I going to write about?)
Popularity Top Ten Lists
The top 10 most linked-to domains on the Internet (at the time of writing) are:
• Gooale.com
• Adobe.com
• Yahoo.com
• Bloaspot.com
• Wikipedia.org
• YouTube.com
• W3.ora
• Myspace.com
• Wordpress.com
• Microsoft.com
The top 10 most linked-to pages on the Internet (at the time of writing) are:
• http ://word press.org/
• http://www.aooale.com/
• http://www.adobe.com/products/acrobat/readstep2.html
• http://www.miibeian.gov.cn/
• http ://vali dator.w3 .org/check/referer
• http://www.statcounter.com/
• http://jigsaw.w3.org/css-validator/check/referer
• http://www.phpbb.com/
• http://www.vahoo.com/
• http://del.icio.us/post
Source: SEOmoz's Linkscape—Index of the World Wde Wsb
Before I summarize I would like to nip the PageRank discussion in the bud. Google releases its PageRank metric through a browser toolbar. This is not the droid you are looking for. That green bar represents only a very small part of the overall search algorithm.
Not only that, but at any gi\en time, the TbPR (Toolbar PageRank) value you see may be up to 60-90 days older or more, and if s a single-digit representation of whafs probablywsrya long decimal value.
Just because a page has a PageRank of 5 does not mean it will outrank all pages with a PageRank of 4. Keep in mind that major search engines do not want you to reverse engineer their algorithms. As such, publicly releasing a definitive metric for ranking would be idiotic from a business perspective. If there is one thing that Google is not, it's idiotic.
Google makes scraping (automatically requesting and distributing) its PageRank metric difficult To get around the limitations, you need to write a program that requests the metric from Google and identifies itself as the Google Toolbar.
In my opinion, hyperlinks are the most important factor when it comes to ranking web pages. This is the result of them being difficult to manipulate. Modern search engines look at link profiles from many different perspectives and use those relationships to determine rank. The takeaway for you is that time spent earning links is time well spent. In the same way that a rising tide raises all ships, popular domains raise all pages. Likewise, popular pages raise the given domain metrics.
In the next section I want you to take a look into the pesky missing puzzle piece of this chapter: relevancy. I am going to discuss how it interacts with popularity, and I may or may not tell you another fairy tale.
কোন মন্তব্য নেই:
একটি মন্তব্য পোস্ট করুন