Quick and Dirty Usability: Leveraging Google Suggest to Instantly Know Your Users

Adam Fourney and Michael Terry
http://hci.uwaterloo.ca/research/query_log_analysis

Play (26min)

Download: MP4 | MP3

Every second of every day, people use Google to troubleshoot problems and to learn how to accomplish their goals. While Google doesn’t make its search query logs publicly available, Google Suggest can be used to learn the most popular queries for any software.

We systematically mined all of the query suggestions for GIMP, Inkscape, Blender, and Scribus to learn about the primary needs and problems encountered by users of these software applications. As examples, our technique collected ~15,000 common queries for GIMP and ~2500 queries for Inkscape.

In this talk, we will present samples of the most common search queries for these applications, and what they suggest about the software user bases and their needs.