A Step-by-Step Guide to Analyzing Search Queries: A Comprehensive Methodology for Replicable Research

When it comes to understanding how users interact with library search tools, every search query tells a story. This research surfaced stories and insights from 600 classified and analyzed search queries submitted between January and May of 2024 from the main search box on the U-M Library website homepage. It also evaluated the results these users encountered in the Library Search “everything” results view. The goal was to gain a deeper understanding of the frequency distribution, relationships, and characteristics of analyzed search queries to guide future design and development efforts. With advancements in AI, this study also aimed to streamline and automate query analysis by experimenting with tools like ChatGPT. Given limited resources, personnel, and time, the research sought to develop a scalable methodology to enable more frequent evaluation of search queries. This paper provides a framework that other individuals and teams can replicate or adapt to better understand trends in user search behavior. This article includes a step-by-step guide to a pilot methodology, insights from the findings, and reflections on observed outcomes. All tools and templates are open-access, allowing others to gather and sample data, classify and analyze search query data, and adapt the processes to specific needs.