The figure below shows the top five search engine results obtained by searching using the keywords “Kelley Blue Book” As can be seen, the results consist of a mix between “link hub” results whose document intent is to serve as a starting point for further navigation (home pages, directories, etc), and results whose document intent is to facilitate a database query (in this case, to find the price of a car of a specific make/model/year).

The Clueray-sorted results for the same search show only documents whose design intent matches that specified by the user: results for the document intent “learn” consist of article-like documents designed to convey information, results for the document intent “interact” show forms and database query interfaces, and Clueray results for the document intent “explore” show homepage and directory-like documents.

Without Clueray, a user interested in finding something other than the Kelley Blue Book home page would have to manually sort through multiple results to find the first document which meets his/her information need. With Clueray, all of the high-quality documents which best meet the user’s information need are automatically promoted to the top of the list.

Note also that the quality of the documents improves from right to left – the better structured articles, forms and directory-like documents appear first followed by others in decreasing order.