Collaborative Filtering is a method or presenting content based on a prediction that if the user likes A, they will probably like B.

Collaborative filtering often works as a recommendation device, for example: at the bottom of a blog post about “Repairing Gutters” you could include links to an article about “Roof Maintenance.” This often presents itself as a series of 3-5 links under a section titled “You may also like.”

Content Management Systems (CMS) can help with collaborative filtering through the use of tags and categories. Similar articles should be tagged with the same keywords or placed in the same category, and the CMS can provide a mechanism for presenting those articles as options for the reader. However, automated systems can be less effective than classification by humans.

Analysis of user paths through your website can predict the likelihood that a user will like other pieces of content. If 60% of your visitors to a Page A proceed to your Catalog page, then it makes sense that you should include a prominent link to that page on the Page A.

Collaborative filtering can be an effective method for increasing Time On Page and conversion rate.