Start free trial Book a demo

Using search analytics for improving content

Speaker

Selvaraaju Murugesan

Head of Data Science, Kovai.co

Duration

34 mins

Presentation

Download Deck

Considering the rapid adoption of Generative AI (GenAI) technologies like ChatGPT and Gemini, users now expect quick and accurate answers to their questions. Traditional knowledge bases often rely on lexical search based on “keywords,” but this is increasingly being replaced by more advanced ChatGPT-like search tools.

These tools will deliver a more accurate response based on the prompt given by the user. So, it’s crucial to capture the intent behind user queries through prompts. By analyzing these prompts, technical writers can gain valuable insights into user intent, identify knowledge gaps, and enhance specific content in the knowledge base.

This webinar by Selvaraaju Murugesan will introduce techniques for utilizing analytics from prompts and combining them with keyword-based searches to improve your knowledge base.

Key takeaways

Era of search engines:

  • Traditionally, search engines were used to find information by typing in keywords. The search engine ranked links based on content quality and its algorithm.
  • With the emergence of tools like ChatGPT and Gemini, people prefer typing questions (prompts) and getting instant answers. When customers land on your knowledge base, they expect a similar experience.

Lexical vs Semantic search:

  • Lexical search: Users type in nouns or adjectives. Google fine-tunes its algorithm using these keywords and produces indexed results quickly.
  • Semantic search: ChatGPT-like interfaces operate on a paradigm of questions (prompts) and provide customized and conversational responses.

Lexical Search Analytics:

  • Lexical search captures search keywords and categorizes them into “Top-performing” and “No results found” keywords.
  • It shows search volumes, trends, and keyword frequencies, revealing knowledge gaps.
  • Technical writers can create new articles or suggest product features based on “No results found” keywords.

GenAI Search Analytics:

  • GenAI-powered search engines like Eddy in Document360 store prompts in the backend.
  • These prompts help technical writers understand user intent better, provide a user-centric response, increase self-service, and improve user satisfaction.

Types of Search Analytics:

  • Topical Analysis: Clusters prompts into general themes, highlighting trending topics and information gaps. Identifies top-cited articles that need regular updates.
  • Usage Analysis: Provides total searches and other statistics, which help to produce business outcome metrics.

Change Management:

  • For a comprehensive view, combine lexical and GenAI search analytics. This approach offers deeper insights into shifting customer behavior and helps identify the information they seek, improving content strategy and customer satisfaction.

About the Speaker

Selvaraaju Murugesan received the B.Eng. degree in Mechatronics Engineering (Gold medalist) from Anna University in 2004 and the M.Eng. degree from LaTrobe University, Australia, in 2008. He has received his Ph.D. degree in Computational mathematics, LaTrobe University. He is currently working as a Head of Data Science at SaaS startup Kovai.co. His interests are in the areas of Business Strategy, Data Analytics, Artificial Intelligence, and Technical Documentation.