Link analysis systems and PageRank is the Google Search’s seventh ranking signal. If you have yet to read about the previously discussed Google ranking signals, you should first read about them!
Here is the list of previous Google ranking signals/factors:
- BERT Explained: Bidirectional Encoder Representations from Transformers
- Crisis Information Systems – CIS
- Deduplication Systems
- Exact Match Domain System
- Freshness Systems
- Helpful Content System
Link analysis systems and PageRank are both exciting topics. You might have read about PageRank in the past, but you have yet to hear or read in detail about the link analysis system.
The fact is that the link analysis system is older than the PageRank system. I won’t mind saying that L.A.S. is as old as the search engines are, even older than Google!
PageRank does the link analysis, but it doesn’t mean that P.R. was the first algorithm on the internet to start doing link analysis.
There was an algorithm even before Google’s PageRank algorithm, which Google has never purchased (As we all know, “G” is habitual of doing this since “don’t know when” 🤫 but not in the case of that older algorithm).
That algorithm is still breathing in another search engine, which I’ll discuss later. 😉
History of Link Analysis System
When the internet started, there were no search engines but directories like DMOZ (old SEO geeks must know about it and must have used it.)
The purpose of directories was to replicate the offline yellow pages and directories, as people were used to searching for things in those books. So, to move them to online search, with the same habit, the directories concept was used.
The second reason for using simple directories was that there were few websites online for which one should have developed any complicated engine. But it all got changed quite fast.
In 1987 there was a search engine called Archie (archie.icm.edu.pl). Archie is the first ever search engine in the world (internet world). That search engine was used to search files placed on FTP servers. For example, if someone has turned on his FTP server and has placed some files in it, Archie was used to searching those files.
In September 1993, Jughead was developed by Rhett Jones at the University of Utah. Even this could have been a better search engine because Jughead used it to search the files in the catalogs of online web servers. It was also working on a directory-like concept.
In November 1993, another search engine was named Aliweb, and YES, it was a good search engine because Aliweb was used to search the web based on keywords and descriptions. Aliweb also provides the facility for webmasters to submit their websites, keywords, and descriptions.
Those old-school search engines were used to rank the web pages based on the following:
- TF-IDF (Term Frequency–Inverse Document Frequency)
- H1’s, H2’s, etc.
- Meta-Keywords (e.g. <meta name=”keywords” content=”link analysis system, pagerank, yaaver”>)
The only problem with their ranking strategy as it was straightforward to manipulate these ranking signals. Because SEO experts were used to:
- Do keywords stuffing in paragraphs
- Use of keywords in invisible text (display: none;)
- Keywords in headings
And search engines had to rank them because they needed to be more intelligent to tackle those manipulations.
Undoubtedly, the internet had become much more prominent in the 1990s. Still, because of no proper search engine algorithms, people couldn’t reach every website online because all websites weren’t indexed/recorded by the search engines.
In 1996, Link Analysis System brought the solution to that problem.
Rankdex algorithm was used to decide the rank of a page based on “how many pages are linking back to that page.” It was the world’s first algorithm, which had started assigning ranks to pages based on their backlinks, showing the pages in search results based on the Rankdex algorithm.
After Rankdex, Larry Page from Google developed his algorithm, which took a step further in the link analysis system. Larry claimed it is insufficient to rank a page based on its total backlinks because Rankdex can be manipulated by creating an enormous amount of backlinks by using backlink automation tools.
(I remember I had done that once and increased P.R. but later, the website got sandboxed 😬. It took me a long time to recover, and I never did that again, of course!)
Larry Page’s algorithm was used to rank the pages based on the quality of backlinks and not only on the bases of the number of backlinks.
Therefore, the ranking system was initially developed by Robin Li Yanhong (Rankdex), but the later one, Larry Page’s PageRank system, was much better than that.
Therefore it got a lot of fame in the internet world and became the primary website rank measurement signal. And that led Google to become the most powerful search engine online.
Wait that doesn’t mean that Robin Li Yanhong’s Rankdex algorithm fainted to death. In 2000, Robin Li designed his search engine named Baidu (I’m sure you guys must have heard this name.)
Baidu is the top-ranked search engine in China.
Baidu was developed before Google. Therefore, there is no harm in saying that Baidu initially introduced the link analysis concept. Google just presented that concept in a much better way. That is the reason Google is “Google!” (period!)
It is enough with the history of the link analysis systems algorithm and PageRank. Let’s get into the formula of Google’s PageRank system and more.
What is the formula of the PageRank system?
The initial PageRank formula/equation was:
Image source: Wikipedia
If you need to improve in mathematics, don’t worry, it is easy to understand this formula. I’ll explain how!
PageRank algorithm decides how much authority or rank should be transferred to another page from one page.
In the formula, “p” is the page that Google ranks.
In the formula, “q” is the page that links to the page “p.” That means “q” is that page where you have created the backlink for your page “p.”
In that Google’s page rank formula, if we divide the total page rank of “q” by all of its outbound links and then we multiply it with a constant value “c,” then the result that we’ll get will be considered as the page rank of our page “p.”
This page rank formula is not the final formula because the constant “c” is a damping factor in this formula. Because we have to add the value of “c.” To explain this damping factor, first, I’ll have to explain what is “Random Surfer Model.”
Google’s Random Surfer Model Explained
Random Surfer Model evaluates the number of chances of a random visitor to visit a particular page. It also determines that the visitor has visited that page by clicking on a link on another web page or has landed directly by typing the URL of that page in the web browser’s address bar. Random Surfer Model shows the results in graphical form.
The page that Random Surfer Model visits the most has the chance to rank higher in results and achieve a higher Page Rank.
Google’s Damping Factor Explained
The damping Factor is a percentage-based value that allows the Random Surfer to move from one group of links cluster to another group of links cluster without following any link between the clusters within the same niche.
The Damping Factor helps Random Surfer avoid getting stuck in a single group of linking webpages or a links cluster.
It helps Google bots explore one webpage along with its backlinks and visit other web pages with a different backlinking group in the same niche to assign the page rank.
From the above discussion, it is concluded that two factors are essential for Google to decide about the page rank of a specific page. The factors are Random Surfer Model and Damping Factor.
Over time, Google has added many other ranking factors to give rank to a page, but Page Rank is the factor that Google has always considered for ranking a web page.
That said, it is also proven that “Backlinks are important for ranking a web page.” After website content, the second important factor for ranking a website is the quality of backlinks.
Therefore, you should first write quality content that follows E.A.T (Expertise, Authoritativeness, and Trustworthiness). and then work on getting high-quality backlinks from well-known websites by outreaching them.
As the Random Surfer Model often visits a webpage, the chances of ranking that page increase.
Your webpage must be getting backlinks from how many pages. Still, at the same time, it is also crucial that your backlinking websites get how many backlinks from other websites along with “how much traffic are those backlinking websites getting.”
Always get backlinks from those pages which are highly connected in themselves and are deeply related to your niche. Get backlinks from sites that already have your website’s related content available because that proves niche relevancy to Google that you have been linked from a similar niche website.
For reference: Here is the list of → All Most Important Google Ranking Signals for 2023.