Findings of a User Study of Automatically Generated Personas


We report findings and implications from a semi-naturalistic user study of a system for Automatic Persona Generation (APG) using large-scale audience data of an organization’s social media channels conducted at the workplace of a major international corporation. Thirteen participants from a range of positions within the company engaged with the system in a use case scenario. We employed a variety of data collection methods, including mouse tracking and survey data, analyzing the data with a mixed method approach. Results show that having an interactive system may aid in keeping personas at the forefront while making customer-centric decisions and indicate that data-driven personas fulfill information needs of decision makers by mixing personas and numerical data. The findings have implications for the design of persona systems and the use of online analytics data to better understand users and customers.

Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems (CHI)