Anais
Resumo do trabalho
Artigos Aplicados · Marketing
Título
SEGMENTATION STRATEGIES FOR POLITICAL MARKETING ON SOCIAL MEDIA: A cluster analysis
Palavras-chave
Voter segmentation
Digital Political Marketing
Clustering in Data Science
Autores
-
VINICIUS CARVALHOCENTRO UNIVERSITÁRIO DO INSTITUTO DE EDUCAÇÃO SUPERIOR DE BRASÍLIA - IESB (IESB)
-
Simone de Araújo Góes AssisCentro Universitário IESB - Instituição de Ensino Superior de Brasília
-
SERGIO CORTESCENTRO UNIVERSITÁRIO DO INSTITUTO DE EDUCAÇÃO SUPERIOR DE BRASÍLIA - IESB (IESB)
Resumo
Introdução
The rise of digital social media has reshaped political communication, establishing an interactive many-to-many model. Castells (1999) describes this transformation as the emergence of the 'network society', in which power flows through informational networks. Campaigns such as those of Obama and Bolsonaro illustrate the strategic use of these platforms. In this context, traditional methods of voter segmentation become insufficient, calling for more dynamic approaches that account for the complexity of political profiles and motivations in the digital environment.
Contexto Investigado
The digitalization of the public sphere has reshaped the logic of political communication, demanding more sophisticated approaches to voter segmentation. In this context, clustering techniques applied to behavioral data from social media offer a promising methodological path to understanding voter profiles more accurately, surpassing the limitations of traditional typologies based on demographic and geographic variables.
Diagnóstico da Situação-Problema
Traditional segmentation in digital political marketing proves limited when faced with the complexity of behaviors on social media. The heterogeneity of the electorate demands methods capable of identifying non-obvious patterns. In this context, the following research question emerges: how can the application of clustering techniques to behavioral data extracted from social media contribute to enhancing segmentation in digital political marketing?
Intervenção Proposta
This study aims to enhance segmentation strategies in digital political marketing by applying clustering techniques to behavioral social media data. It draws on benefit segmentation theory (Haley, 1968), digital political flows (Castells, 1999), and microtargeting practices (Zuboff, 2019). Using a quantitative and deductive approach, it applies the AutoCluster framework to Instagram data from Brazilian governors, analyzing engagement, frequency, and communicational structure.
Resultados Obtidos
The analysis revealed three digital profiles among the brazilian governors: the "institutional manager," characterized by consistent activity and moderate engagement; the "efficient mobilizer," marked by a high proportional interaction rate; and the "mass communicator," distinguished by a high volume of posts and significant reach. Clustering proved effective in replacing fixed categories with behavioral patterns. It is recommended to expand the analysis to other platforms and incorporate NLP to deepen the qualitative understanding of the electorate.
Contribuição Tecnológica-Social
The research advances the field of political marketing segmentation by integrating data science and cluster analysis based on digital behavior, overcoming traditional models rooted in demographic and geographic profiles. It contributes a replicable methodological framework applicable to real-world campaigns in social media, promoting communication personalization and data-driven ethical decision-making. Socially, it strengthens strategic intelligence in the public sphere. Technologically, it innovates by automating segmentation through AutoCluster and multivariate analysis on social networks.