Uncovering Search Secrets From Google Patents — 1

During my free time, I love to go through Google Search Patents partly because of the fact that they’re very interesting to read.

Most of the time, I happen to gather some valuable insights regarding how search works and how websites get ranked. So, I decided to share some of those insights with you.

Let’s make it very clear that it’s not necessary that Google is going to incorporate all of its search patents into ranking algorithms but going through them will give us a broader picture of how ranking algorithms work and to what extent they can go to provide search users with top quality results. And SEO executives! you know why this is important for you.

The last Google Search Patent I went through was “Reasonable Surfer Patent”. I found that very interesting so let’s talk about it in this post.

Ranking Documents Based on User Behaviour and/or Feature Data

Here’s what its abstract reads

A system generates a model based on feature data relating to different features of a link from a linking document to a linked document and user behavior data relating to navigational actions associated with the link. The system also assigns a rank to a document based on the model.

This patent talks about a model that employs methods to assign a weight to the link in the referencing webpage based on feature and navigational data associated with the link. Accordingly, it assigns a rank to the referenced webpage based on the weight of the link in the referencing webpage.

This also talks about assigning a rank to the linked page based on the ranking of the linking page.

The weight to the link is based on user behavior data and some feature data associated with it. User behavior data may include things such as navigational action such as clicks on the link.

One of the interesting things I found was that feature data associated with the link (pointing to some other page) might include the context of words around the hyperlink and topical cluster surrounding the link besides including some other things such as length and font of the anchor text etc.

Feature data associated with the linking page might include the topical cluster it is associated with, degree of match between the topical cluster of the page and the topical cluster of the anchor text, etc.

User behavior data might include the link a user clicked among a group of links, language of the user, queries performed by the user, etc.

The weight calculated for a particular link based on the feature and user behavior data associated with the link and the linking page is used to calculate the probability of the link being clicked and followed by the user.Then the weight of the link is used to assign a ranking to the linked page.

To sum up, In order to get an idea about the weight of a particular link in the webpage, ask yourself this simple question: How likely is someone going to click on it? If the answer is yes, then the link is valuable and is helping with the ranking of the linked page. But, if the answer is no, you might want to change its position, anchor text, font, etc.

What Should You Do?

Ask your content writers to surround the hyperlink with topically relevant words. This is important since this helps search algorithms gauge the relation between the referencing and referenced webpage.

If there are multiple links on your page, consider moving the important ones up so that user finds and follows it. This helps with the ranking of the linked webpage.

Provide a highly descriptive and short anchor text.

Insert the link in your blog in such a way that it makes readers highly likely to follow it. If there’s a link in your blog that no one clicks then it doesn’t serve any purpose there.

Use Page Analytics chrome extension to check how individual links are performing, in terms of clicks, on your webpage.