Rankbrain is a machine learning algorithm that allows Google to offer highly pertinent responses to search queries, including searches that it has never seen before. This approach is based on training and repeated crawls to attempt to classify web pages by content theme, and then to rank them for queries for that content. The ranking is then adjusted based on user behavior: do searchers click on the result? Do they click, but then return to the search results?
How to optimize your site for Rankbrain
The Rankbrain algorithm takes into account the content of the page, even if the user’s search keyword does not appear on the page, as well as how a user interacts with the page. To optimize for Rankbrain, simple steps include:
- Ensure that the Rankbrain algorithm can understand the content of your page by using named entities and structured data
- Link to thematically related pages to reinforce the subject of the content
- Provide engaging content that keeps users on the page
Oncrawl and Rankbrain
Oncrawl’s metrics allow you to find your pain points where Rankbrain is concerned and to create KPIs that track your performance with this type of indexation.
- Use Oncrawl’s log analysis to track Google crawl behavior on key pages
- Check your use of named entities using Oncrawl’s SEO Toolbox
- Verify the distribution of structured data on your site with Oncrawl
- Use Oncrawl’s semantic analysis to verify that the n-grams on a given page cover concepts closely related to the main content