The new frontier of customer understanding: Financial satisfaction and AutoML in banking

The objective of this study is to examine the factors associated with financial satisfaction using microdata from the Bank of Spain's Financial Competence Survey (2016 and 2021). Machine learning allows us to explore a wide range of possible predictors of financial satisfaction, overcoming the limitations of traditional approaches. The proposed methodological design combines predictive accuracy with interpretability, capturing hierarchical effects and interaction patterns that traditional approaches often overlook. The results consistently highlight household income and employment status as key structural factors, while perceived financial constraints, ability to meet expenses, and financial concerns emerge as influential perceptual correlates. The comparative analysis reveals two complementary segmentation logics: one dominated by objective resources and another focused on perceptions and planning behaviors, with evidence of a shift between 2016 and 2021 toward the growing importance of subjective assessments. This approach enables us to identify the variables most strongly associated with financial satisfaction, complementing existing models in the literature. Substantively, the results highlight that financial well-being is not only related to objective resources but also to how individuals interpret and manage them. These associations should be understood as correlational patterns rather than causal effects. From a managerial perspective, the findings underscore the need for retail financial institutions to integrate perceptual and behavioral dimensions into segmentation strategies to anticipate dissatisfaction, tailor services, and strengthen long-term consumer relationships.