Introducing a Model of Automated Brand-Generated Content in an Era of Computational Advertising

Abstract

Advancements in computing, technology, and their applications to advertising enable marketers to deliver brand messages tailored to individuals and consumer segments. The growth of computational advertising (CA) has created new opportunities but also poses risks in the use of algorithms to generate and optimize the impact of such messages. This article addresses a particular domain influenced by these advancements, namely, automated brand-generated content. We offer an automated brand-generated content (ABC) model that posits two advances. First, rather than solely optimizing consumer data for enhanced impact of automated content, we submit, and provide extra key variables to further illustrate, that there is a desirable balance of both consumer and brand data as inputs to algorithms to serve short-and long-term impact goals. Second, this article guides research by addressing tensions between understanding the relationship between inputs and desired impacts (both short and long term) and proposing a research agenda for future work.

Using AI predicted personality to enhance advertising effectiveness

“…Secondly, the study demonstrates that AI is a powerful tool in the profiling of consumer personalities. We build on an emergent body of research, bringing together the psychology and marketing literature (Schwartz et al , 2013; Plank and Hovy, 2015; Arnoux et al , 2017; van Noort et al , 2020), to establish AI as a reliable means of personality profiling, facilitating the comprehensive analysis of vast quantities of data, to generate insights pertinent for marketing purposes. Previous research has examined personality traits within the continuum of the same trait (Margolis et al , 2020; Moon, 2002).…”

“…Previously, the creation of advertising was the jurisdiction of designated personnel, yet as we enter an era of computational advertising, the automated generation of content that draws on consumer behavioural data to personalise marketing messages, is pervasive (van Noort et al , 2020). However, as of yet, research in marketing is yet to apply AI technology to determine personality traits.…”

“…In addition to academic research, technology vendors such as IBM have developed software to predict personality which uses natural language processing to understand personality traits (IBM, 2018). AI technology can infer personality characteristics from text-based information using an open-vocabulary approach (Schwartz et al, 2013;Plank and Hovy, 2015;Arnoux et al, 2017;van Noort et al, 2020). An emerging stream of research identifies that language usage patterns reflect the personality, thinking style, social connections and emotional states (Hirsh and Peterson, 2009;Yarkoni, 2010;Fast and Funder, 2008).…”

“…The application of AI in personality research is burgeoning, albeit emergent, with improvements in natural language processing together with the growth of digital communications providing an abundance of consumer behavioural data (Davenport et al , 2020; Diehl et al , 2003; Kietzmann et al , 2018; Flavián et al , 2019; van Noort et al , 2020; Helberger et al , 2020). Empirical applications include AI processing of data captured using an eye tracker which suggests that eye movements during everyday activities can predict personality (Hoppe et al , 2018) and AI prediction of individual personality traits using brain activity, heart rate, sweating and pupil dilation in response to images and videos (Berkovsky et al , 2019).…”

“…AI technology can infer personality characteristics from text-based information using an open-vocabulary approach (Schwartz et al , 2013; Plank and Hovy, 2015; Arnoux et al , 2017; van Noort et al , 2020). An emerging stream of research identifies that language usage patterns reflect the personality, thinking style, social connections and emotional states (Hirsh and Peterson, 2009; Yarkoni, 2010; Fast and Funder, 2008).…”