Publicação
Customer lifetime value in insurance
| Resumo: | Throughout the years, companies from several business sectors have strived to strengthen their client portfolio by acquiring and retaining the most profitable. For this to happen, current and potential clients must be clearly classified based on their past and future interactions with a company throughout the lifetime of their relationship. This report presents how the previous scenario was implemented using Customer Lifetime Value (CLV) in one of the biggest bancassurance companies in Portugal, during a 9-month internship. Before delving into the detailed set of this project phases, the concept of CLV was reviewed, as well as the characteristics which define its several approaches, followed by the alignment of the chosen approach to the company reality. This CLV model was limited to a 12-month future horizon, covered 7 company dimensions (one global, plus 1 per lines of business) and took into consideration as main future client interactions churn, cross-sell, upsell and risk of claiming. These previous components were modeled with the help of SAS Enterprise Miner or estimated using SAS Enterprise Guide and analyzing historical events. Besides a purely monetary CLV, it was also generated an ordinal output using a set of business rules and a ranking data discretization method. Finally, a back-test validation procedure was executed to evaluate the reliability of both types of outputs in each of the considered dimensions and its results were analyzed. |
|---|---|
| Autores principais: | Abreu, Jorge Eduardo Carvalho |
| Assunto: | Customer Lifetime Value CLV Customer Current Value Upsell Cross-sell Churn Risk Insurance Bancassurance Valor actual do cliente Risco Seguros |
| Ano: | 2019 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | Throughout the years, companies from several business sectors have strived to strengthen their client portfolio by acquiring and retaining the most profitable. For this to happen, current and potential clients must be clearly classified based on their past and future interactions with a company throughout the lifetime of their relationship. This report presents how the previous scenario was implemented using Customer Lifetime Value (CLV) in one of the biggest bancassurance companies in Portugal, during a 9-month internship. Before delving into the detailed set of this project phases, the concept of CLV was reviewed, as well as the characteristics which define its several approaches, followed by the alignment of the chosen approach to the company reality. This CLV model was limited to a 12-month future horizon, covered 7 company dimensions (one global, plus 1 per lines of business) and took into consideration as main future client interactions churn, cross-sell, upsell and risk of claiming. These previous components were modeled with the help of SAS Enterprise Miner or estimated using SAS Enterprise Guide and analyzing historical events. Besides a purely monetary CLV, it was also generated an ordinal output using a set of business rules and a ranking data discretization method. Finally, a back-test validation procedure was executed to evaluate the reliability of both types of outputs in each of the considered dimensions and its results were analyzed. |
|---|