Capturing the change in topical interests of personas over time

Abstract

In this research, we collect monthly content consumption and demographic data from YouTube over two years for a large media publisher. We use automation to generate 15 personas each month and examine the consistency of the generated personas over time. We find that there are 35 unique personas in total for the entire period, reflecting the changes in the underlying audience population. For each persona, we generate topics of interest and identify the top three monthly topics for each of the 35 personas following an identical algorithmic approach each month. We then compare the sets of topical interests of the personas month‐over‐month for the entire two‐year period. Findings show that there is an average 20.2% change in topical interests and that 68% of the personas experience more topical change than topical consistency. Findings suggest that the topical interests of online audiences are fluid and changes in the underlying audience data can occur within a relatively short period, resulting in the need for constant updating of personas using data‐driven methods. The implications for organizations seeking to understand their online audience are that they should employ routine data analysis to detect changes in the audience interests and investigate ways to automate their persona generation processes.

How categorization impacts the design of requests: Asking for email addresses in call-centre interactions

“…Pruitt & Adlin 2006). For example, Jansen, Jung, & Salminen (2019:127) describe how they collect demographic data and information about the content consumed by YouTube users to generate personas and analyse their stability and consistency over time.…”

A Survey of 15 Years of Data-Driven Persona Development

“…For MPD, updating personas requires excessive amounts of costly and non-scalable manual labor, resulting in the personas often not being updated at all. DD PD can capture the change of user behavior over time (Jansen et al, 2019;Jung et al, 2019), as it relies on automated processes for periodic data collection and easy re-analysis using standard algorithms.…”

“…In the third period, researchers discover social media and online analytics data for persona development Jansen et al, 2019;Salminen, Jung et al, 2019b). Also, "data science algorithms," such as matrix factorization (An et al, 2017), are applied for persona development through frameworks, such as Python's scikit-learn.…”

Data-Assisted Persona Construction Using Social Media Data

“…The challenge for data-driven persona construction is to monitor and identify how change happens over time: the veracity, velocity and volume. Findings show that topical interests, as reflected by personas constructed using data from online sources, change by an average of over 20%, while only a third of the personas in those cases experience topical consistency [12]. This shows the necessity for a constant update of the personas in order to reflect the changes in topical interests.…”