| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 4.78 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Este trabalho tem como principal objetivo estudar o comportamento dos sócios do Sport Lisboa
e Benfica e traçar os perfis encontrados, através de uma análise de grupos. O trabalho foi realizado
através de um estágio na Direção Comercial e Marketing do Sport Lisboa e Benfica, mais
concretamente no departamento de CRM. São usados dados de várias bases de dados do SLB,
incluindo fontes não só do departamento de CRM, mas também de outros departamentos, como
os departamentos de Sócios, Merchandising, Bilhética, entre outros. É descrita a forma como os
dados foram obtidos, transformados e carregados para uma base de dados, construída no Power
BI, de modo a conseguir-se aplicar um modelo capaz de analisar os padrões comportamentais dos
sócios. Na descrição do tratamento de dados, são abordados os métodos para imputar missing
values. Através do R Studio são explorados e testados os algoritmos de agrupamento, juntamente
com as diferentes técnicas de agrupamento, tendo sido usado o agrupamento hierárquico.
Finalmente, são utilizadas técnicas de visualização, com recurso ao R Studio e Power BI, para
estudar o comportamento dos sócios de cada grupo, comparando-os, e destacando os traços que
melhor definem os seus perfis. Com esta caracterização dos sócios por grupos é possível, por
exemplo, executar campanhas de Marketing dirigidas.
The main aim of this work is to study the behaviour of Sport Lisboa e Benfica’s members, defining customers profiles found, through a cluster analysis. The work was carried out through an internship in the Commercial and Marketing Department of Sport Lisboa e Benfica, more specifically in the CRM Department. Data from various SLB databases is used, including sources not only from the CRM Department, but also from other departments, such as Membership, Merchandising, Ticketing, among others. It is described how the data was obtained, transformed, and loaded to a dataset, built in Power BI, in order to apply a model capable of analysing members’ behavioural patterns. In the description of the data processing, the methods for imputing missing values are discussed. The clustering algorithms are explored and tested using R Studio, along with the different linkage techniques, using hierarchical clustering. Finally, visualization techniques are used, using R Studio and Power BI, to study the members’ behaviour of each group, comparing them and highlighting the traits that best define their profiles. With this members’ characterisation by groups, it is possible to, for instance, carry out targeted marketing campaigns.
The main aim of this work is to study the behaviour of Sport Lisboa e Benfica’s members, defining customers profiles found, through a cluster analysis. The work was carried out through an internship in the Commercial and Marketing Department of Sport Lisboa e Benfica, more specifically in the CRM Department. Data from various SLB databases is used, including sources not only from the CRM Department, but also from other departments, such as Membership, Merchandising, Ticketing, among others. It is described how the data was obtained, transformed, and loaded to a dataset, built in Power BI, in order to apply a model capable of analysing members’ behavioural patterns. In the description of the data processing, the methods for imputing missing values are discussed. The clustering algorithms are explored and tested using R Studio, along with the different linkage techniques, using hierarchical clustering. Finally, visualization techniques are used, using R Studio and Power BI, to study the members’ behaviour of each group, comparing them and highlighting the traits that best define their profiles. With this members’ characterisation by groups, it is possible to, for instance, carry out targeted marketing campaigns.
Descrição
Mestrado Bolonha em Métodos Quantitativos para a Decisão Económica e Empresarial
Palavras-chave
Análise de grupos CRM Comportamento do consumidor Agrupamento hierárquico Método de Ward Cluster analysis Customer behaviour Hierarchical clustering Ward’s Method
Contexto Educativo
Citação
Valente, Diogo Monge (2023). “Sport Lisboa e Benfica : análise do perfil e comportamento do sócio”. Dissertação de Mestrado. Universidade de Lisboa. Instituto Superior de Economia e Gestão
Editora
Instituto Superior de Economia e Gestão
