Learn

There is no question that website traffic from Search Engines is an important marketing consideration. And links from other websites are considered an integral part of a comprehensive Search Engine Optimization (SEO) strategy.

Those who research how Search Engines rank website pages will tell you that good SEO is a combination of 3 primary ingredients:

1. On-page factors
2. Off-page factors
3. Page Rank (Google); Web Rank (Yahoo)

On-page factors are those attributes of a specific website page. The most important are keyword density found in body, title and description tags. Google claims to evaluate 130 attributes of every pages but it is unlikely that all these attributes influence their Search Engine Results Page (SERP).

But all the important search engines have acknowledged that webmasters can manipulate their on-page factors to an extent that makes this ingredient a secondary factor in their ranking algorithm.

Off-page factors are the links that appear on other website pages and are considered by many as the most important ingredient for 2 reasons.

1. They are considered a 3^rd party reference of your website.
2. Links on some one else's website are difficult to manipulate.

Page Rank is a quantifiable method of determining the interconnection of any one website page to other website pages. The more connections or links, the higher the Page Rank score. This quantifiable measure is patented by Google and named after one of the Google founders, Mr. Lawrence Page.

Yahoo claims to have a similar measure called Web Rank available on their toolbar, but at the time of this writing, it is not functional and little is known about how the number is calculated.

The rest of this document is devoted to a discussion of Off-page factors, their importance, construction, reputation and relevancy.

Link importance -

As discussed briefly above, links that reside on other websites are considered a very important ingredient for all the major search engines. That is because there are lots of information implied in a link and links that reside on another website are difficult to attain and manipulate.

A link to your website is often described as a vote for your website. The assumption is that other webmasters are very particular to whom they link.  Webmaster prefer linking to websites that they believe their visitors would find interesting and want to be associated with.

Since links are under the control of another webmaster, search engines consider them less prone to manipulation and compared to on-page factors, a better indicator of what a website page is really about.

But this does not mean that off-page factors are completely free of manipulation. In fact, the necessity of links for search engine rankings has birthed a cottage industry. Link exchange businesses that will attain links from other websites on your behalf are now common. Some webmasters have created multiple websites for the expressed purpose of linking these sites together, thus, forming a link ring. Others have created directory websites for exchanging links known as "link farms" or "free link communities".

But there is another problem that the search engines don't talk about too much. Sometimes a legitimate website might link to another website with or without knowledge that the other has adult content. Search engines may rank the legitimate website well for its keywords. But if the legitimate website has an outbound link to an adult content site, visitors may find themselves on a website page they did not bargain for. These searchers hold the search engine accountable for ranking a website that is linked or 'associated with' an adult content site.

Search Engines know that the SEO community understands that links are important and have developed creative ways to give the Search Engines what they want.  So it is not surprising that Search Engines have deployed and are continuing to develop new methods of filtering out "junk links" (defined by GoogleGuy) and sites containing links to adult content websites.

Link construction -

There are 2 attributes of every link.

1. The visible portion: either a graphic or text
2. The href portion: html code containing a URL that points to another website page, also known as the target page.

Links that are used in advertisements are most often constructed with affiliate URLs. These URLs commonly have tracking codes ('?' and/or '=' characters or long numeric strings) that Search Engines detect. (example found on MSN:>http://g.msn.com/0AD0000Q/611135.1??PID=2124716&UIT=G&TargetID=1001195&AN=25713&PG=INVIHS). And since these types of links are not considered unbiased, Search Engines do not count them.

Links that do not have tracking codes or variables in the URL are referred to as 'static URLs'. Search engines count these links but the value of the link depends on the visible portion.

Graphic links are pictures that may contain symbol or words, but in either case, the search engines do not understand what the graphic may be saying. They cannot read graphics. So there is no additional value in graphic links.

Visible portion		Href portion
		Affiliate URL	Static URL
Graphic Link		No Value	Some value, counts toward Page Rank
Text Links	URL		Better value
	Words		Best value





Text links give the search engines something to read. The words that are clickable are directly associated with and anchored to the href portion, thus the name 'anchor text'. Search engines even have a search command dedicated to anchor text. In Google, simply type the search command 'allinanchor:keyword' (replace keyword with a specific keyword phrase; eg. allinanchor:dog) will return website pages that have links whose anchor-text matches the keyword (presumably most to least). Yahoo's command is slightly different: 'inurl:keyword'.

Search engines associate the anchor-text of a link with the target page. As the number of links with specific anchor-text increases, 'link reputation' is formed.

Link reputation -

Link reputation is defined by the anchor-text associated with links. The more link reputation a website page has, the stronger its rankings in the SERPs (particularly in Google).

It is this simple principle that has lead many webmasters to choose domain names that have keywords in the domain separated by "-" or "_" marks. If a domain name (e.g. www.flower-baskets.com) is used as the link anchor-text and the domain name has your primary keywords in it, then your link reputation will increase. This only works if your domain name contains your keywords. Note that keywords in your domain name must be separated by "-" or "_" marks, otherwise, the search engines will not be able to detect distinct words in the domain name. Likewise, if your primary keywords are "Botanical buckets", a domain name like www.flower-baskets.com will not contribute to your desired reputation.

Link relevancy -

Search Engines assume that websites with the highest number of strong reputation links from other related websites must be the 'authorities' and deserve better rankings. Yet, Search Engines do not currently detect if a link is from a related website.

Up until mid 2003, a link was a link - no matter where it was from. As long as the search engines crawled and indexed a page with a link on it, and the link had good reputation, it was a good link. So some webmasters exploited this 'loop hole' and got links from non-related websites. GoogleGuy calls these types of links 'junk links'. Now, Search Engines are deploying and developing methods to filter out 'junk links'.

Although still disputed, Google appears to be filtering links based on class C IP address. The premise for this filter is based on Google's 2003 patent.

Just because something is mentioned in a patent does not mean the patent holder has to employ the methods. But observational evidence does seem to support the theorem. It appears that Google may be either filtering or at least de-rating links from the same class C IP address.

IP filtering does make sense. First, the method can be used to either filter or de-rate the value of links on the same domain. Second, the method can be used to detect multiple links from another website.

Lastly, links that reside on the same class C IP would necessarily be on the same host and have a high probability of being a related website (mirror, clone, or site authored by the same webmaster / company). This filter would thwart efforts of webmasters that construct a number of websites, host all the sites on the same host provider, then link them all together to boost their link reputation.  It would not affect links from websites that are hosted on different hosting providers. But this latter approach is a lot more painful and may have technical issues if each website is served from a common database.

Google has also created a website 'black list'. This black list appears to be made up of websites that have violated Google's website policies. The list is not available to the public so it is hard to know who is 'naughty' and who is 'nice'. Several webmasters have reported cases of website penalizations when a legitimate website has a link to a black listed site.

But in order to determine if a link is from a relevant website, search engines need to do more. Google has already started to roll out Topic Sensitive Page Rank (TSPR). Other search engines may use a form of website classifications.

Whether TSPR or website classifications, both share a common technical obstacle: Search Engines must figure out a way to determine the meaning of a search phrase and relate the phrase to some form of classification. For example, should a link from the Java tourist bureau website count as a relevant link to a website about Java scripting?

If you type "java" as a search term, should the search engine return websites about:

• the island named Java (a regional category)?
• coffee (a lifestyle category)?
• a programming script (computer category)?

Consider the search term "web ring". Are links from nature sites relevant? What about "closed cell"? Are links from websites about our judicial system relevant? Perhaps the user is referring to special research about living cells – or a terrorist cell?

Many of the words we use have several meanings. It is context that determines the meaning or 'sense' of specific words. Perhaps this is the reason 'personalized search' is such a hot topic these days. For it is 'personalized search' that relates keywords with website classifications, and website classifications allow the search engines to qualify related links.

Google is already knee deep into TSPR. Google has launched 2 beta projects:

• Personalized web search ( http://labs.google.com/personalized )
• Flavored Search functionality ( http://www.google.com/services/siteflavored.html )

Each of these beta programs was introduced in the first ½ of 2004 and gives us a glimpse of the power of TSPR. In order to understand how TSPR works, we need a little background on PR (Page Rank).

The equation for PageRank is:

PR(A) = (1-d) + d(PR(t1)/C(t1) + ... + PR(tn)/C(tn))

Where:

't1 - tn' are pages linking to ‘A’

'C' is the number of outbound links on that page

'd' is a damping factor, typically set to 0.85.

In order to run the equation, you must know all the pages linking to 'A' and the PageRank of each of these linking pages. The only way to perform this calculation is to iterate: run the equation many times, typically 20-40 times. In order to determine the true PageRank of any page in Google's index, you must be Google: have an index of how all website pages are connected to each other.

In laymen's terms, PageRank is a measure of the interconnection or popularity of any one website page as measured by links from other website pages. It is commonly thought of as 'votes'. The more links or 'votes' from other website pages, the higher the PageRank score. But there is one more element to the equation. The value of a 'vote' from a website page is divided by the number of outbound links (total 'votes'). So then links or 'votes' from pages that have fewer outbound links cast more of their voting power to a page.

The important thing to note is the use of the damping factor of 0.85. We don't really know the exact value that Google uses, but we do know that there must be a value here (perhaps +/- 0.05). If we assume that 0.85 is the actual value, then we can say that Page Rank only accounts for 85% of the possible value. So what about the 15% remaining? It really does not matter much if all website pages are calculated in the same manner. But what is important is that there is a 15% component that can be added to each website page and this component can be the TS (Topic Sensitive) portion of TSPR. Another way of saying it is that 85% of a calculated value is PR, 15% is TS.

So how is TS figured out? You can read the original authors paper here: http://dbpubs.stanford.edu:8090/pub/2002-6, or read on for a more simplified explanation.

The theory of Topic Sensitivity starts with the assumption that there are authority websites for a specific subject or keyword phrase. These authority websites link out to other websites and other websites link to more websites. Every time there is an outbound link, there is a component of that link that had its origins from the authority websites. The topic sensitivity that is passed through the link is defined by where the upstream websites got their links.

It may be easier to think of Topic Sensitivity as a 'bloodline'. Some dogs are pure breeds, but most dogs are mutts, a combination of bloodlines. Mutts may have blood from a few or many different dog types just as websites have links from different website categories. If a dog has a lot of bloodline from golden retrievers, then presumably that dog would be better at hunting than other dogs. Likewise, if links to a website are primarily health related, then the website would rank better for health related keyword phrases.

Therefore, prior to the mid 2003 era, Google rankings were largely dependant on what others said you were (link reputation). In the new era, your rankings will be dependant on who is saying what you are (link relevancy and link reputation).

Yahoo and MSN do not have TSPR, but they can employ similar schemes. Their strategy could involve the classification of websites and the correlation of keyword phrases with each category. Then Yahoo and MSN could determine which links to your website are relevant to your subject matter and filter out the benefits of non-relevant links.

This may sound easy enough but the implementation of TSPR or website classification strategies are extremely complex. First consider that every keyword phrase must be classified. Then every website page must be classified. Now do this across 4+ BILLION website pages that are continually being modified/changed/added and deleted and deliver search engine result pages (SERPs) in less then 0.25 seconds. The classification task alone is enormous requiring extensive computing power and the most advanced semantics algorithms (the study of words and their meanings).

Google appears to be way ahead in this process. They have already created beta programs that can be used to seed their keyword classification process. It is likely that Google is allowing the public to use these beta programs so that they can classify keyword phrases in real time. If the data is statistically significant, this process can be used to loop back into their Topic Sensitivity algorithm. The more it is used, the better it gets.

Other search engines may need to wait for advances in computer technology and semantics. The best semantics algorithms have an accuracy of about 40% compared to the 'golden standard'. This is not a very encouraging value but it is far better than results just 2 years ago; so much better that industry is starting to apply significant resources.

Even with semantics advances, systems must be scaleable. 64 bit microprocessors are well suited to semantics calculations and will have a significant impact on calculation speeds over current 32 bit microprocessors. In coming years, Search Engines will be able to take advantage of lower cost 64 bit microprocessors and semantic algorithm advances that will directly affect SERP quality.