In any Google SEO effort, one of the most vital elements that is outside the control of any SEO Expert is the age of your domain name or website. Most people are not aware that the age of your website is an vital aspect of Google’s PageRank algorithm. Our testing has found that using the best Search Engine Optimization practices there is a limit that will be reached as to the keywords that we can reach top 10 results in. For example, if you are optimizing your website for keywords that have a monthly search volume of over 10,000 searches, you should expect to be able to reach top 10 organic search results no sooner than 4-6 months after you register your web site and start your SEO efforts.
We have tested these numbers and have arrived at the following conclusions. Google won’t rank new websites for keywords that get very large monthly search volumes for the simple reason that it has to give an advantage to the older, more established websites that have endured the longest. This really makes significance. With the ever-more marketing nature of the internet and the enormous number of sites that are simply produced to exploit user information from visitors, and how quickly they come and go, the number 1 measure of how relevant a website is can only be single-minded by their age.
Keep in mind that an ancient website that continually changes focus and topics will also be tagged as a new website, this practice is to avoid websites who have legitimately gained ranking from using their web status to then promote marketing harvest or air force.
Lesson to learn here: plot ahead, focus your site with excellent plotting ahead of time, so that it only requires small changes during your SEO strategy and not major topic changes.
Posts Tagged ‘Algorithm’
Website Age – An Important Part Of The Google Pagerank Algorithm
Saturday, March 6th, 2010Google PageRank Algorithm
Thursday, February 25th, 2010Reading How Google Finds Your Needle in the Web’s Haystack I was surprised by the simplicity of the math underlying the google PageRank algorithm, and the ease with which it seemed to be efficiently implementable. Being able to do a google-style ranking seems useful for a wide range of cases, and since I had wanted to take a look at python for numerics for some time, I chose to give it a shot (note that there by now exists a python implementation using the numarray module).
Contents:
Note that PageRank™ is a brand of Google and that the described algorithm is covered by numerous patents. I hope the powers that be will nevertheless ignore/forgive my insolence — I was just curious and mean no harm 
. Note also, as multiple commenters have pointed out, that this algorithm is an implementation of the “historical” PageRank algorithm as in print by Page and Brin, which does not really seem to be actively used at Google anymore.
A recent article on the AMS featurecolumn neatly described the google PageRank algorithm, showing in a few relatively simple steps that for a set of N linked pages it boils down to finding the N-dimensional (column) vector I of “page relevance coefficients” which one can formulate to be (see the above article) the principal eigenvector (i.e. the one corresponding to the largest eigenvalue) of the google matrix G certain as
G = ?(H + A) + (1 – ?) 1N . (1)
Here, ? is some parameter between 0 and 1 (typically taken to be 0.85). For the entry specified by row i and column j we define Hij = 1 / lj if page j links to page i (and lj is the total number of links on page j), and 0 otherwise, such that the “relevance” of page i is
Ii = ?j Hij Ij , (2)
corresponding to the number of links pointing to each page, weighted by the relevance of the source page divided by the number of links emanating from the source page. Similarly, we define Aij = 1 / N if j is a page with no outgoing links, and 0 otherwise. So each page has a total of 1 outgoing “link weights”, i.e. the sum over the elements of each column of H + A is one (it is a stochastic matrix). Irrevocably, 1N is certain to be an N x N matrix with all elements copy to 1 / N (it is not the identity matrix), and is therefore also stochastic. Similarly, G is stochastic. This latter statement is vital, because for stochastic matrices the largest eigenvalue is 1, and the corresponding eigenvector can accordingly be found using the power method, which will turn out to be very well-organized in this case.
Summarizing, G (a finite markov chain) may be interpreted to model the behaviour of a user who stays at each page for the same amount of time, then either (with probability ?) randomly clicks a link on this page (or goes to a random page if no outgoing links exist), or picks a random page off of the www (with probability 1 – ?).
Irrevocably, the power method relies on the fact that for eigenvalue problems where the largest eigenvalue is non-degenerate and copy to 1 (which is the case), one can find a excellent approximate for the principal eigenvector via an iterative procedure early from some (illogical) guess I0:
I = Ik = GkI0 , k?? . (3)
Full Article: Google PageRank Algorithm
The Layman’s Guide to the Google Pagerank Algorithm
Tuesday, January 12th, 2010Let me question you: If you have a certain thought or topic in mind, and you wish to find out more about this topic, what do you do? Ten years ago you would probably have gone to the library, but today… You GOOGLE IT!
If you take anyone currently living in the modern world, chances are that is what they will tell you. Google is King! Over the course of just a few years, Google has gone from a couple of smart guys at Stanford University with the revolutionary thought of making the entire internet available from their desktop, to being the undisputed gatekeeper to nearly every single part of humanity’s collective knowledge.
The Google search engine has in fact become so standard and proliferous that the word ‘google’ itself has now become a verb! (As in, if you want to find out more about a certain person, you just google them.)
Every time you do a search, you will see the term or phrase that you searched at the top, followed by about ten webpages that Google thinks are most relevant to your term. Since there are literally tens of millions of Google searches every single day, it is not a fantastic leap to reckon that the websites that manage to get their ranking very high in Google will get ALOT of free visitors and traffic.
But what is it exactly that determines which websites get listed in the top ten listings? Well, the system that is behind every single search result that you see is called the PageRank system, named after its creator and co-founder of Google, Larry Page.
Previous to the PageRank system, there did exist some other methodologies for determining web search relevance and delivering accurate results, but none of them were as robust, accurate, democratic, or resistant to human error as PageRank.
-What Puts PageRank in A League of Its Own-
There are in the end two major thoughts behind the PageRank system that have made it so revolutionary:
First, the PageRank system is rather democratic in nature because every time one website (we will call it site A) links to some different website (we will call it site B), that link is considered to be a ‘vote’ by site A that site B has excellent information, or for some reason is worthy of being viewed and read. This concept of the democratic nature of the links found all over the internet is a vital main thought behind the PR system.
Second (and this is the part that really place PageRank on the amount), NOT ALL LINKS ARE CREATED EQUAL!
That is to say that if you have two links coming to your website, one from Forbes.com and a further from some backwater, glide-by-night dot com, these two links will not be treated equally.
So what does this mean for the question of how did the highest ranked sites get where they are? They have been around for long enough to have numerous standard sites link to them, they have valuable, relevant, dynamic content, and chances are that they probably link to other related websites.
A further vital (but not so revolutionary) mechanism behind determining which webpages are showed for certain keywords is an advanced text-matching system. Google’s text-matching system is able to deliver highly relevant webpages because of the vast computing power behind the Google search engine itself.
-Technical Explanation of a Website’s PageRank-
This following part is a technical explanation for those who want to further know the nature of the PageRank algorithm. If you are only interested in learning how to improve your own site’s PR, then feel free to skip to the next section.
With the PageRank algorithm, every single website on the internet is given a numerical PR value somewhere between 1 and 10, with 10 being the best. It will help if you can remember from your math class what a logarithm is, because the assignment of a certain PR number is logarithmic in nature, similar to the Richter scale of measuring earthquakes.
This is vital to know, especially if you want to increase your own PageRank. In terms of PageRank, this means that a PR6 site is not twice as valuable as a PR5 site, but really TEN TIMES as valuble. So this would mean (not exactly but approximately) that an incoming link from a PR6 site would give you as much value as about a few dozen PR4 sites. Notice that a PR6 incoming link will NOT give the value of 100 PR4 links, because PageRank is concerned with the quantity of incoming links as well as how vital they are.
-Tips For Improving The PR of Your Site or Blog-
Try to make content that is valuable, pun, or for some reason really makes people want to link to it. This strategy will increase your number of incoming links without any extra work, which will so increase the PageRank of your site or blog.
A ‘link farm’ is a website that has hundreds or thousands of incoming and outgoing links. Sites like this can actively inflate PageRank to make a site seem more relevant than it really is, so Google will ‘punish’ websites associated with link farms by bringing them down in the search rankings.
Do not worry or feel like your site or blog is not excellent enough if after just a few months or so you do not have a high PageRank and are not listed very high in the search results. It takes time to build PR, so the better your content and the longer you have been online, the better chance you have at genuinely gaining a higher PR.
See if you can find a few high-feature websites or blogs out there related to your own site topic, and contact the owner to see is they would be interested in between to your site if you link to theirs. This is called ‘link exchanging,’ and if you do it to much then Google may ‘punish’ you becase this is a further way of inflating PR, but exchanging links with a few feature sites will help you.
One last thing, and this has been stressed right through the article, there really is a single golden rule that you can apply to boost your PageRank: make MASSIVELY VALUABLE information and content that people will genuinely want to link to on their own, and you are set.