Block Determination and Link Distribution

Google is getting better and better at recognizing what blocks within your Web page serve what purpose. For Google this is especially effective when trying to figure out what the main content of your website is, but it is also a good way to see what content is prominent for the visitor and what isn’t. Some blocks of content might even be hidden by default and only interaction caught by JavaScript will make it visible.

Only recently, Google has been getting quite good at calculating how JavaScript and Stylesheets alter the way texts and images are shown. This changes a lot in the methods available to make certain content stand out more and also has a lot of influence on the link value that is distributed from page to page. The main conclusion? “Links within your main content have become much more important!”


The Function of a Block

Many Web pages distinguish between a header, a footer, sidebars, titles, headings, and the main content of the page. A visitor that scans all these blocks easily selects what is worth reading, but for a machine that is slightly harder. An increasing amount of SEO research shows that Google does evaluate all these blocks differently. Earlier tests indicated that Google’s detection was probably based on the duplicate content of these blocks, but that they probably didn’t differentiate between the role of each block. Now we have stacking proof that a header can be treated significantly different from the footer and it is important to know how this impacts the way you optimize your SEO.

The Location of a Link

An old-fashioned practice of making links less prominent for visitors was to put them in the footer at the bottom of your page. This was for instance used as the entrance to a cluster of SEO landing pages.

Because Google will now also deem those links as less important, it serves no purpose anymore. Rather, interlink your pages from within the relevant and unique content and don’t create pages just for SEO purposes. Integrate all important keywords within pages that have a logical role within your website and choose important pages for the more heavily competed keyword combinations. Important pages are from now on the ones that a visitor is likely to navigate to and not just the ones heavily linked to from your footer menu.

This way, the distribution of link value within your website has become less predictable. It will probably never become as easy as PageRank calculations again, where the amount of links determined how much value each page will receive.

Visibility and Prominence

Functionalities like endless scrolling, drop-down menus, and help texts require interaction and JavaScript to appear for visitors. Content not located in the HTML, but only available through JavaScript, seems to be indexed by Google, so how important does the Googlebot find such texts? For humans the main question would be, how logical is the interaction that triggers the appearance of the content? Is Google smart enough to guess that?

Because still so much is unknown, it is advisable to still have an HTML version of all content available. You should at least help Googlebot by using the right HTML markup for your content blocks and, if applicable, and HTML 5 indicators for content functionality. Google seems to look at visual prominence, but examples with header tags like H1 on a page are still not beaten in ranking by examples with a paragraph that looks exactly the same. Just make sure the tags you use do not conflict with the signals JavaScript and CSS could give to Google.

Google wanting to become more like a human visitor is probably a good thing. It definitely opens a lot of new challenges, but also creates a lot of new weaknesses that spammers can outsmart. Let’s hope this “Pinocchio update” doesn’t bring more problems than it solves.

Related reading

Photograph of people brainstorming around a table.
A hand holding a transparent piece of plastic or glass, with the Google logo superimposed onto it.

Simple Share Buttons