Generating Personas for Products on Social Media: A Mixed Method to Analyze Online Users
Hao Tan, Sheng-Lan Peng, Jiaxin Liu, Chun-Peng Zhu, Fan Zhou · International Journal of Human-Computer Interaction · 2021
The purpose of this research is to develop a methodology that combines the quantitative and the qualitative analysis to generate personas for products on social platforms. The user data on social platforms contain massive information relating to the lifestyle people have and the products people use or are interested. By analyzing the specific content generated by users on social platforms, e.g., content involving the term “tablet,” it is possible to reveal how the users consider or use the product, “tablet.” By analyzing the users’ homepages, the information relating to the users’ daily life can be found. We collected 276, 675 pieces of relevant data regarding the product, “tablet,” from 12, 965 online users on China’s widely used social media platforms. Then automatic user segments and the profiles of each group were generated and structured by natural language processing technology. The results of these quantitative analyses were then qualitatively examined by manual analyses, which provide additional insights and detailed descriptions on the automatically generated persona profiles. In this study, six personas representing distinct user types were created. The mixed method of combining the quantitative and qualitative methods makes the generation of personas faster and more insightful. The generated personas can represent real user behavior and characteristics and can provide insights into the products, which also can provide support on designing new products and optimizing existing products.