Uma abordagem multidimensional para a análise da experiência do consumidor no setor automotivo

Abstract

In the automotive sector, understanding customer profiles at different moments of interaction with the company (before, during and after the purchase) is essential for the development of new products and to provide a satisfactory experience with the brand. However, the diversity of contexts, moments, and profiles generates a combinatorial and interdependent problem that transcends the informational power of existing computational solutions, which are generally onedimensional and focused on visualization. This article proposes a multidimensional approach that captures imperceptible nuances of the interaction between customers and companies in the automotive sector and expands the informational power as a competitive differentiator. The approach, presented in the form of a computational tool, allows integrating heterogeneous data and reconfiguring, in a short space of time, a large set of consumer and product analysis scenarios. In addition, the approach integrates statistical and machine learning techniques, such as hierarchical agglomerative clustering, to segment consumers and identify the main differences between groups, which can be valuable from a business management perspective. The methodology is validated with real data from Renault do Brasil and can be extended to other brands in the sector.