What are the most important Google Ranking Factors?

SEO’s are looking at Google rankings as an output of algorithms and not so often they are trying to go deeper into the Google algorithms.
We can do this by reading and analyzing Google patents. By doing this we as professional SEO’s will understand much better the output – Google rankings – but starting with the input – Google Signals -.

But wait. Are all these Ranking Factors or Ranking Signals?

People in the SEO world are using two terms when it comes to Rankings: Signals and Factors.

I consider that both terms are fair to use just the perspective of looking into these is different.

• Google engineers may refer to these inputs as ranking “Signals”.
• I consider that SEO’s are referring to these inputs as ranking “Factors”.

That being said, let’s dive right in and find out more details for a few of the most important Ranking Factors.

Domain Age and Rate of Linking

This section will reveal why having backlinks is so important.

But before having more details, Google tells us that is considering the age of a web page as the first time they see that web page as referenced in another website or web page.
Also, Google patent tells us about how Google may look the links pointed to a site, calculate the average rate of links pointed to a site and use that info to rank a site, based upon that linking.

Creating valuable content that people will find good enough to read and share with other peers. But this is not something new and we heard in hundreds of article this advice.

But when it comes to the types of pages we usually consider to point links to, what are they? Are these product pages where the only part that is pure content is the part of the product description? No, clearly no!

What Google patent is telling us about links:

Ideally, a search engine, in response to a given user’s search query, will provide the user with the most relevant results. One category of search engines identifies relevant documents based on a comparison of the search query terms to the words contained in the documents. Another category of search engines identifies relevant documents using factors other than, or in addition to, the presence of the search query terms in the documents. One such search engine uses information associated with links to or from the documents to determine the relative importance of the documents.

Google patent is also talking about different other components when it comes to links and why they may consider creating a “ranking score”.

Consistent with aspects of the invention, the ranking score is a value, that attempts to quantify the quality of web pages.

Google may use the inception date of the web page for the scoring of the web page.

The patent is also introducing the concept of fairly recent inception date and the fact at the inception date the website will not have a significant number of links from other web pages – backlinks -.

Velocity Rate or Link Growth Rate

Google is introducing the concept of growth link rate.

Consider the example of a web page with an inception date of yesterday that is referenced by 10 backlinks. This web page may be scored higher by Google than a web page with an inception date of 10 years ago that is referenced by 100 backlinks because the rate of link growth for the former is relatively higher than the latter.

But Google also admits that there could also lower the score of the page to reduce the effect of spamming. But that rate if is a good rate can be used to score the page, “giving more weight to documents to which links are generated more often“.

That’s why an ongoing link-building campaign is so important, that’s why building high-quality links continuously are important.

The concept of “link-based” score is used:

H=L/log(F+2)

• H – history adjusted link score.
  L – give link score given to the page – using another patent formula U.S. Pat. No. 6,285,999 -.
  F – elapsed time measured from the inception date associated with the document.

For some queries, OLDER documents may be more favorable than newer ones.

Link-based criteria

The patent section indicates that several new links and several lost links could count as a formula in identifying the behavior of links pointing to a page.
The same section is coming also with the concept of links are weighted differently.

Raise or Lower the Score based on Freshness of a Link

Google says that may raise the score of a page based on how fresh or not the links pointing to that page are.
The scoring may be employed recursively.
Let’s have a quick example.

Page S is 2 years old. Page S is considered fresh IF n% of the links to S are fresh
OR if the pages containing forward links to S are considered fresh.

Anchor Text

The patent is referring to the scenario when an anchor text changes over time, and the score of the page that gets a backlink is changing as well.
A very important detail in regards to the anchor texts is when the text of a page changes significantly or when the anchor text changes significantly.

Google may consider to IGNORE (or discontinue) all links and/or anchor text prior to that date.

Traffic

Information related to traffic with a document over time may be used to generate or ALTER a score associated with a page.

A large reduction in traffic may indicate that a document may be stale – no longer be updated, or may be superseded by another document -.

Google may compare the AVERAGE traffic for a page over the last J days – j=30 – or adjusted over the last K days – k= 365 -.

Google can adjust the page scoring, by tracking the traffic patterns.

Looking to these patterns Google is also considering the “advertising traffic” for a particular web page.

• Has the page advertisement?
• What is the quality of advertisers?
• What is the CTR on the page to the advertisers?

Google may use these “time-varying” characteristics relating to advertising traffic to score the document.

User Behavior

Google may monitor the number of times that a document is selected – click – from a set of search results.
An example is worth it to identify how Google will consider improving the ranking score for a specific page.

Let’s assume that your travel blog article about “family travel packages” is ranking #6. Google is measuring the time visitors are spending on your article and create an average time that is using as a reference.

Along with other ranking factors, your page now is ranking on #3 for the same keyword.
Google measures again the time visitors spent on your article and compares it with the reference that has been already stored.

Now, the visitors are spending less time on your page, due to some banners you added to that page to monetize your article better.

But soon, your page based on the user behavior factor will lose that #3 ranking since Google will consider scoring poorly
your web page accordingly to the new time spent by users to your page.

Domain-related Information

Many time the SEO’s are coming with a piece of information that it’s difficult to be verified. They are based on personal experience or maybe based on rumors.
Studying Google patents is a quite new practice and that was initiated by a well-appreciated professional SEO, Bill Slawski.

One of the ranking factors that Google is considering is the time a domain name has been registered.

A 10 years registered domain name it tells Google that the domain name is legit is not there to deceive search engines.

Google it include in this category also the DNS for a domain that may be monitored to predict whether a domain is legitimate or not.

Ranking History

Google may monitor the time-varying ranking of a web page in response to search queries.
Google may determine that a page that jumps in rankings across many queries might be topical or it could signal an attempt to spam.

The quantity of rate that a page moves in rankings over some time might be used to influence future scores.

A top result may receive a score of 1.0, down to a score near to 0 for the Nth result.

Google is using ranking history not only to monitor and track page rankings but also to determine when a page is ranking for a commercial query.

Google also may see that a commercial query has a higher chance to be spammed so may treat the pages associated with those queries accordingly.

The history of rankings for a page on a given query is not the only metric that Google can monitor, but few others correlated to that respective page:

• Number of queries for which a page is ranking
• Rankings fluctuations (increasing/decreasing)
• Seasonality
• Changes in its scores over time for a URL-query pair

The section of ranking history is also dealing with a score that Google can create by
monitoring a number of pages as an “independent query-based criteria” over time.

The main purpose of doing this is to create an average score for a set of queries that
can be used later on as an indicator of a “hot topic” or other particular phenomena.

We are often using in our competitive backlinks analysis a certain metric of “links velocity”.

As well Google is tracking a ranking spike that can trigger a signal of spamming Google rankings for that particular query.

• Google may take measures to prevent spam attempts, by employing different signals to allow a rank to grow at a certain rate.

Google clearly shows that is using different mechanisms to fight against spam attempts, that for sure are not rare cases especially for the top commercial queries for high competitive
verticals. But it also admits that a certain high authoritative websites (e.g. Yahoo) may be considered as exceptions.

Google also admits that may not consider as a spammy technique a higher number of links pointing to an authoritative website as being spam.

But a very important detail is revealed at this point, indirectly. There is also a “threshold” that Google may consider for ranking growth of a page.

Also, significant ranking drops could be considered if a page is outdated.

Looking at those elements in this Google rank history patent section, I consider that Google it uses a “self-reinforcing loop” that is acting as a stand-alone signal.

…If the rank of a page drops significantly over time, then Google may consider the web page as outdated and score the page accordingly.

Just to make my statement more clear, Google is generating or alter – depends on case by case – a SCORE for a given page, before rank that given page on a certain rank for a certain query.

User Maintained / Generated Data

Let’s define first, how Google define”generated data”.

“Bookmarks”, “favorites”, or other types of data that may provide some indication of pages favored by or of interest to the user.

Many times I heard SEO’s talking about Google’s tracking user’s data and behavior from Chrome. It’s obvious that this is happening, but just a few offers also the evidence that this is happening.

Now, we have the facts and the facts are Google patents.

Google may obtain this data either directly – via a browser assistant – or indirectly – via a browser -. Google may then analyze over time several bookmarks/favorites to which a page is associated
to determine the importance of the page.

Several users accessing that page from their bookmark/favorite list is important because Google may consider that behavior as an indication that the page is relatively important.

This type of signal is another score that Google uses in its ranking algorithms.

But Google is going even further with this type of data they collect, analyze and use as part of users behavior as a ranking score of a certain page.

They may consider also the temp or browser cache files, cookies associated with a page might be monitored and used to determine whether there is an upward or downward trend in interest for a web page.

Unique Words & Phrases in Anchor Text

Professional SEO’s are using a word when it comes to the backlinks and anchor text.

“Exact match” or “phrase match”.
Mostly these terms are referring to the anchor texts are using an exact match of a high-value commercial query to improve rankings for that specific keyword.

Google is also using this signal in identifying the number of identical anchors but with the clear goal of identifying spamming web pages.

What is interesting at this point is how Google engineers are thinking in terms of anchor texts. By using an algorithm they may cap the impact of suspect anchor texts but also they may use a continuous scale for the likelihood of “synthetic” anchors.

A score is generated and that could improve or alter rankings based on the unique anchor texts used, phrases used as anchor texts.

Making the things a bit more complicated for the readers – not with this goal in mind – Google may consider also using bigram frequency in the English language in order to identify if the anchor texts and phrases use a plain English language.

Linkage of Independent Peers

Google has developed in this section a mechanism that is also monitoring the incoming and outgoing number of links and it triggers especially when that number indicates a sudden growth.

A large number of links to individual pages may indicate a potential attempt to spam. Going deeper than that Google is also correlating the number of links to the number of anchor texts.

Is this mechanism the core Google algorithm in fighting with spammy techniques?

Demoting links is a post-Penguin algorithm implementation.
Demoting the impact of a high number of links especially when a link-based scoring technique is used is considered
by Google.

There are two implementations of demoting links:

• Binary decision item (e.g. demote the score of a web page by a fixed amount of links).
• Multiplicative factor (I personally consider this implementation is used within the competitive verticals where we know historically that a high volume of links is acquired).

Document Topics

When it comes to the topicality of an article or of a web page, Google has developed implementations are above the scope of this article.

What is important at this point and in regards to the ranking factors is how Google is looking to identify if a web page, the article is relevant or not for a certain category and what ranking score is going to acquire overall.

Some factors are used by Google in regards to generate or alter a score of topicality associated with a web page, article or website section:

• Categorization
• URL analysis
• Content analysis
• Clustering
• Summarization
• Unique low frequent words
• Monitoring the topic of a web page over time

Monitoring of the topic is an important factor since could be used in both ways. Improving the score but also alter the score of a web page or article if the original topic has changed.

Conclusion

There are many other signals/factors and lot’s of articles are dealing with the subject. The purpose of this article is just to offer a different perspective of how Google in their patents is dealing with the ranking factors.

My goal as a professional SEO is not just to analyze the output – web pages and their rankings – but to look deeply into the Google’s patents and to analyze what elements they put into the algorithms.
I only scratched the surface of what Google’s published in their patents, but I will continue to publish more articles dealing with “the source” of rankings factors.

The SEO industry is well-diversified and SEO’s as the name of a famous and very popular SEO book “The Art of SEO” implies are “artists”. I’m trying to be as much as possible an engineer and less an artist.

I would love to hear your thoughts about the Google patents and what term do you prefer: SEO Ranking Signals or Ranking Factors.