What is PageRank?

In the past, PageRank was official Google data that indicated a website’s "popularity" rating, or rather a page’s popularity. Indeed, PageRank could vary from page to page within the same website. It was found on the "Google PageRank" toolbar which displayed a score from 0 to 10, defined according to an algorithm. This toolbar disappeared in 2016 and has not been replaced. That year, there was turmoil among SEOs who wondered if the PageRank notion still existed. Today, we can safely say that the answer is: yes and no!
Well, that’s very helpful!
In reality, PageRank is inescapable because it’s at the heart of all Google's algorithms. Without this notion, how could Google define which pages to index and position? It’s just no longer accessible to the public.
In addition, this is a criterion that has become much more complex than before because it takes into account a large number of factors from different algorithms. Back in the day, a page's value was measured by the number of links it received. But in the face of the proliferation of poor-quality directories, doorway pages and other link buying platforms, Google has had to crack down on it. Today, backlinks’ quality has become just as important as the quantity of backlinks, if not more important.

How does Google PageRank work?

PageRank is simply a mathematical formula, which looks like this:

PR(A) = (1-d) + d (PR(T1)/C(T1) + ... + PR(Tn)/C(Tn))

Okay, that was just a little info but it may not be the formula that will help you understand how it works.

Google takes three factors into account when calculating a page’s PageRank:

  • Inbound links quantity and quality
  • Number of outbound links
  • PageRank for each link on the page

Let's take pages A, B and C.

  • Page C receives 2 links: one from page A, and one from page B
  • Page A has a higher PageRank than page B, and fewer outbound links.

Enter this information into the PageRank algorithm, and you will get page C’s PageRank.

The PageRank formula also incorporates a "damping factor" which simulates the probability that a user will continue to click on links while browsing the web. Each click’s importance decreases with each link.

In other words, the likelihood that someone will click on a link on the first page they visit is high. But then, the probability that he/she clicks on a link on the next page is slightly lower... and so on.

Since Google needs several pages’ PageRank to calculate that of a single page, one might wonder how it comes to calculate the initial page’s PageRank?

Here is an extract from the original article on PageRank:
PageRank, or PR(A), can be calculated using a simple iterative algorithm and corresponds to the main eigenvector of the web's normalized link matrix.

In other words, PageRank can be calculated even without knowing the linked pages’ PageRank because it’s a relative score, and not an absolute. A page’s quality is also assessed against the other pages, a bit like teachers who, upon finishing grading homework, adjust the marks according to the class’ overall level.

How to measure PageRank?

Therefore, we have just seen that the PageRank notion still exists, but that the official tool for learning about it is no longer available. Unfortunately, since this is data that only Google owns, there is no PageRank toolbar replica. But it’s possible to get an estimate.

Ahrefs URL Rating

URL Rating is a metric that indicates a URL backlink profile’s strength (from 0 to 100).

It’s comparable to the original Google PageRank formula because it:

  • Takes nofollow attributes into account
  • Integrates the damping factor
  • Takes into account all the URLS found

However, while Ahrefs' URL Rating gives us a good clue about a page's popularity, it can't be used as the only official criteria. Google's PageRank has evolved considerably over the years, and today it incorporates very advanced criteria, secretly kept by Google. This fact alone makes it very difficult to understand how Ahrefs' rating differs from Google's current iteration of PageRank.

TrustFlow and CitationFlow on Majestic

Trust Flow is a measure that analyzes a website’s reliability based on the links’ theme and quality. This metric is determined by the number of clicks a specific page receives from a set of websites.
A website's trust index will increase if it’s authoritative and qualitative. This means that the TF score will be determined by the amount of traffic generated by the link, and the website’s relevance from which the link originated.
Like Trust Flow, Citation Flow measures a link's popularity, by referring to the number of clicks the link receives. The only difference is that it ignores links quality. In other words, if a website has a higher Citation Flow than Trust Flow, it probably means that it’s receiving a lot of low-quality links. An ideal link profile must therefore display a perfect balance in Trust Flow and Citation Flow, which represents a very condensed point cloud on the Majestic graph.

By mixing quantity and quality data, Majestic's metrics therefore tend to approximate PageRank criteria.

Moz's Domain Authority and Page Authority

The Domain Authority (DA) is a metric that predicts how well a website will rank in the SERP (Search Engine Results Page). It varies from 1 to 100. As with the URL Rating, it’s calculated based on several criteria similar to those of PageRank, with the difference that it gives an overall score, for the entire domain. Moz also offers the Page Authority (PA) score to assess a single page’s relevance.
Moz's DA and PA are accessible from the free Mozbar Chrome extension, the Link Analysis Tool, the Keywords Explorer’s SERP Analysis section, and many other SEO tools.

The LinkJuice, by SEMJuice

Some specialists who would like to go further have developed their own classification criteria. The Link Juice developed by SEMJuice is among the most relevant to date. It mixes together many metrics from Majestic, SemRush, Monitorank and many more, while applying a clever weighting according to each criterion’s importance.
It’s possible to know your Link Juice by registering on the platform, which is intended primarily to help you improve your popularity through netlinking.

How to improve your PageRank?

To preserve and improve a PageRank, you have to act on 3 levers: internal linking, outbound links and backlinks.

The internal mesh

The way pages relate to each other affects a website's PageRank. This is because the "SEO juice" that is sent to a website from external links is often sent to the homepage. The homepage is a website’s level zero. Through internal links, the homepage transmits the popularity it receives to other deeper pages. An optimized internal linking helps this SEO juice to be transmitted to the pages that interest us. This is called internal PageRank. Therefore, internal linking does not improve PageRank, but it does help to keep it level.

Outbound links

To understand this concept, we must keep the notion of “SEO juice” in mind. If a page receives quality juice from a page, it’s in your best interest to keep it. This requires not only limiting internal links to other pages on the website, but also to other websites. Otherwise, the page’s PageRank will be “diluted”.


Inbound links are the biggest source of PageRank, only if they are of good quality. It’s better to have 10 quality backlinks on 10 trusted websites (sharing a common theme), than 1,000 links on directories or content farms. To assess the quality of the links you may have with partners, you can use the metrics available in Majestic (Trust Flow), Moz or Ahrefs.


PageRank still is an essential and important metric in search engine rankings. Despite Google’s total opacity on this topic, it’s still possible to assess a page’s importance thanks to various metrics developed by reputable SEO players. However, PageRank should not be the only data to evaluate a page’s effectiveness. It’s also essential to measure its technical performance with tools like PageSpeed Insights or GTmetrix.

   Article written by Louis Chevant

Further reading

The complete guide to Internal Meshing

The step-by-step method to build your semantic cocoons, your mesh and the optimal tree structure of your website.