Mobile advertising spreading through personal social networks using a viral approach and branded apps

Resumo

Mobile advertising has substantially transformed over the past few years thanks to advancements in mobile technology, wireless networks, software tools, and platforms. This transformation has given rise to a complex mobile advertising ecosystem, which is characterised by multiple agents and platforms that support the new requirements of the market. The complexity of the mobile advertising ecosystem has implications for several key topics, including user privacy, ad spreading, and brand relationship measurement in wireless mobile environments (WMEs). Research from diverse disciplines, including social sciences, computer science, and marketing, has played a significant role in addressing the challenges of mobile advertising, thus contributing to its evolution. Despite these research efforts, several topics in mobile advertising research remain understudied. Specifically, the spread of mobile advertising through users personal social networks (PSNs) and the factors that can influence this spread in WMEs are topics that require further research to enhance the understanding of mobile advertising. This thesis aims to propose a model that facilitates the study of the spread of mobile advertising via users PSNs using a viral approach and branded apps as a channel for interaction. Our model was designed to provide a simplified mobile advertising approach with a direct relationship between advertisers and users. This new model allows the study of several factors that can affect the spread of mobile advertising, such as the users privacy perception, user feedback, and superspreader users. We used simulation as the primary methodology to test the models that we developed to address our research questions. This method allowed us to design hypothetical scenarios and simulate them utilising an agent-based modelling (ABM) tool and the SIR disease propagation model to recreate the spreading of mobile advertising using a viral approach. Through the analysis of simulation outcomes, we gained insights into the behaviour of the variables that are essential to the specific objectives outlined in our thesis. This analysis enabled us to identify key findings, draw meaningful conclusions, and contribute to the mobile advertising domain. Our research findings can be categorised into four main axes associated with the research questions set in this thesis. These axes include the mobile user profile, the brand relationship in WMEs, the spread of mobile advertising and factors that can affect ad propagation. In our study, we proposed a mobile user profile model that successfully identified users PSNs based on mobile phone data, revealing differences between their inferred social circles and face-to-face interactions. Regarding brand relationship research findings, our proposed user-brand relationship (UBR) measurement model simulation indicated that branded app usage time was the most critical factor in determining the strength of this relationship. Moreover, we identified two approaches for generating the number of active users of the branded app, which enabled us to validate our model under hypothetical scenarios. The simulation of our mobile advertising model revealed that the percentage of spreader users of ads and the number of levels in the virtual space were significant drivers in increasing the number of cumulative infected users by ads and the duration of the propagation of mobile advertising. Furthermore, incorporating user privacy perception levels in the model had a positive effect on ad propagation, as mobile users were more likely to accept and share mobile ads with contacts in their PSNs. However, including user feedback in the model had a negative impact on advertising propagation, while incorporating superspreader users had a positive effect when included at the peak of infected users growth by mobile ads in the model simulation. This thesis contributes to expand the literature on mobile advertising in several ways. We provided a simplified advertising model that enables the study of the propagation of mobile advertising through PSNs and using branded apps. We also developed a simulation of this model that combines the advantages of the SIR viral model and ABM for analysing the propagation of mobile advertising and studying specific questions regarding ad spreading. We provided a simplified user profile model that uses a combination of mobile phone data and users social interaction data to generate a mobile user profile. This profile is seamlessly integrated into our new mobile advertising model to provide users PSNs data for ad spreading. We also proposed a model for the measurement of UBR strength that uses a new methodological approach based on specific metrics of branded apps as a gauging context. From a practical perspective, our research provides helpful insights for promotion managers seeking to improve their mobile advertising campaigns by leveraging new methods for ad spreading. Furthermore, managers can experiment with simulated scenarios to test different user privacy perception settings and refine their mobile advertising strategies based on their evaluation.