<p>Promoting the efficient use of water requires in-depth knowledge of the specificities and particularities involved in water consumption dynamics. In this sense, this study aims to evaluate household water consumption using cross-sectional and longitudinal approaches. Data related to construction characteristics, socioeconomic and demographic variables and water consumption habits were collected from a sample of 400 households in Southern Brazil. Correlation and regression analyses were employed to ascertain the main drivers of water consumption. Additionally, the COVID-19 pandemic effect was assessed through analysis of variance (ANOVA) and Prais-Winsten regression model. Multiple linear and geographically weighted regression models demonstrate that per capita water consumption is explained by the number of residents in the household, home ownership, built area, swimming pool presence, and bucket use for washing external areas. The results highlight the spatial variability in demographic variables as well as in the construction characteristics of the households. Regarding the COVID-19 pandemic, the water consumption was higher during the first two waves, returning to previous levels after the end of the pandemic. This study highlighted the importance of a comprehensive analysis, considering time and spatiality, to identify key elements regarding residential water consumption. In-depth knowledge of water consumption in the built environment is of unequivocal importance for integrated urban water management and the results of this research aim to contribute to this end.</p>