Challenge
Bartleby, a popular learning platform, was facing a significant challenge with its search function. Analysis revealed that the search function was highly utilized, with a focus on Q&A solutions. Despite this, the success rate of search was only 50%. The challenge was to improve the relevance and effectiveness of search results to encourage frequent returns, increase user retention, and enhance user engagement.
Solution
A series of solutions were designed to:
Assist users in finding the "best" solution for their search question
The Search Results Logic was enhanced, the Top Result was highlighted, and Metadata Surfacing was improved.
Assist users in finding the content type or Bartleby feature that will answer their question
Content Variety and Feature Discovery were introduced
The main objective was not solely to enhance successful search rates, but also to convert search users into recurring Bartleby users.
Main features
-
The order in which search results are presented to users was refined, ensuring that the most relevant results are displayed first.
-
Duplicates were removed and 'Contains searches' were introduced, enabling users to find results containing specific words or phrases. The top result was determined by several criteria including exact match, historical engagement, and popularity in the user's country.
-
Relevant metadata such as Textbook Information, Video, Topics, and Bartleby Certified were made visible to assist users in quickly assessing the relevance of search results.
-
The variety of content presented in search results was expanded to include Related Concept Explainers and Lit Guides.
-
It was made easier for users to discover features like 'Ask an Expert', 'Live Chat', 'Practice Problems', 'Essay Builder', and 'Math Solver
-
Further enhancements such as Search Preview, grouping results by micro-topic, and search via photo were explored.
The Outcome
By implementing these solutions, the definition of a successful search on Bartleby was redefined. It's now characterized by a user-initiated search that leads to a click on a search result which answers the user's problem and results in them remaining on the solution for a significant amount of time. A/B testing was utilized to compare the effectiveness of the solutions and continuously refine the approach.
