There have been two interesting articles on how Google may be measuring the value of (internal) links on the page:
1. In this article on whether the location of a link matters for SEO, Google’s John Mueller states:
…the location of internal links does not necessarily matter if they are in the body content, footer, header or other sections on the page. The content, the anchor text, of those links helps Google understand what the page is about but the links aren’t really used more than just for crawling your site.
He did note that this might be different for external links.
2. This article by Roger Monti offers a bid different perspective:
Google’s Martin Splitt made it clear that Google divides a web page into sections and that the main section where the important content exists is called the Centerpiece Annotation.
In the article, Martin explains that the main content section is what Google uses to understand what the page is about.
Moreover, Martin said that the different sections are “weighted” differently.”
So the question remains: How are links weighted when there are many of them on the same page?
To discuss this question, we invited Bill Slawski, the well known SEO expert with over 26 years of experience. Bill Slawski is the foremost expert on Google’s patents as related to SEO.
Takeaways from the video:
- The reasonable surfer model is in place, and was updated about 4 years ago via continuation patent. The higher the probability that a link will get clicked, the more PageRank it will pass.
- When link text is not informative (e.g. “click here”) Google takes into account surrounding text to assign meaning to that link (there’s nothing wrong from SEO standpoint in using those links)
- Internal link equity is “limitless”: No need to “sculpt pagerank” by removing or nofollowing internal links on a page: There’s no limit to PageRank that can pass to all of them but Google will assign different amount of link equity based on how clickable it is (and the anchor text map Google creates). None of those links will be ignored even of there are many.
