Last week, Google was granted a patent that describes how Google might segment the content of a web page. Depending on the location of the content, a web page will get higher or lower rankings.
Why is it important to segment a web page into different parts?
The average web page consists of several parts: a header, a footer, the website navigation, the main content, ads, etc.
Not all parts of the page are equally important. To determine the position of a web page in the search results, text that appears in the main content of a web page is more important than text that appears in the footer or in ads.
How does Google analyze the web page content?
According to the patent text, Google looks at the HTML of the page. Google also tries to simulate the display of a web page in a web browser to understand the location and the purpose of different parts of the page.
What exactly does Google analyze in the different parts of a web page?
The patent lists several “modules” that are used to analyze a web page:
Link analysis: Links found in different semantically distinct regions may be assigned different weights. That means that the influence of a link in the main part of a web page is probably higher than the influence of a link in the footer.
Text analysis: The importance of text on a web page is rated by location. For example, if a keyword appears in the main content area of a web page or in the title, then this page will rank higher than a page that has the keyword in the footer.
Image captioning: Text found near an image is more relevant to the image than text further away from it. This module can be used to rank a picture in image search.
Snippet construction: Google usually does not use the meta description tag for the snippet that is displayed on the search result page. According to the patent, Google creates the snippet based on the text in the web page section that is the most relevant to the keyword.