In September 2019, Majestic launched a major upgrade to their link index, including the new concepts of Link Density and Link Context at a block level. By “Block-level” I mean that Majestic has divided every page in its index into up to 40 separate segments, allowing links to be analyzed in the context of the segment or section of the page in which it lies, rather than in the context of the whole page.
This is a BIG step change! It has been under wraps for over a year in the making – but as I am not an employee at Majestic anymore, even I was kept in the dark. Now I know, I put together my first impressions in this video…
Looking a bit into the science, the idea of block-level analysis for search engines is not new, but link tools have not been able to exploit the theory, partly because it requires so much extra computing power and storage.
Now that links can be analyzed in segments in this way, there will be less need to rely on links being measured based on single value metrics for web pages, because we as SEOs can now see exactly what kind of link we are looking at. There may be other factors, such as domain age and the date the content was posted, but on the whole, I found the visualizations (which I understand a patent pending) very easy to understand and I also found the filtering options to be almost everything I could ask for. Links in paragraphs of text? easy. Links above the fold? easy.
Link context is nothing to do with the context of the web page itself and everything to do with the HTML surrounding the link. Until now, the only equivalent was Ahrefs’ “snippet” feature, which contains a few words each side, but Majestic has leapfrogged this idea to include much more context – whether the surrounding HTML contains other links, or images. Whether other links are internal or external, foloow or nofollow. The user can even see and render the extended markup to get a picture from the Majestic dashboard of the link on context.
Link density again does not look at the number of links on the source page itself (although that information remains available). Instead it looks at the surrounding text of the segment and works out what percentage of that immediate text is in links… whether external OR internal. This makes it very easy to see whether a link stands out for the user, or whether it is buried and therefore unlikely to be clicked on.
Reasonable Surfer Proxy
Google are believed to have augmented their famous PageRank algorithm many years ago with the concept of the “reasonable surfer“. This algorithm, I believe, skews the value of each link in the PageRank model, so that the same overall amount of PageRank is distributed from a page, but with a heavy bias towards the links most likely to be clicked on. The problem here is that there is very little research on what a person is MOST likely to click on at a high level. Is it the navigation or the links in blocks of text? At least, now, SEOs can have a practical way of conducting research that may help to surface answers to these kinds of questions.
I think it is a valuable new way of looking at links. I am biased, of course, but so would you be! 🙂
Research cited in this article:
Cai, D., He, X., Wen, J.R. and Ma, W.Y., 2004, July. Block-level link analysis. In Proceedings of the 27th annual international ACM SIGIR conference on Research and development in information retrieval (pp. 440-447). ACM. (Link to paper)
Google’s Reasonable Surfer revisited. Bill Slawski. SEO by the Sea. (Link to article)
How Pagerank Works (Link to video post)
Link Context and Link Density white paper (Majestic Blog)