The development and production of bananas are influenced by spatio-temporal edaphoclimatic conditions, which explains in a certain way the heterogeneous response that occurs on the farms, highlighting the need to generate information regarding the morphological characteristics of fruit for taking corrective actions that entail to increased productivity. The objective of this research was to carry out an analysis of production, taking into account the quantifiable variables through bunches profiling. 98,956 bunches recorded in a period of 108 weeks between 2021 and 2022 were analyzed in ten farms of two banana groups in Urabá-Antioquia, in the municipalities of Turbo, Apartadó and Carepa. Variables such as: cluster age, cluster weight, number of hands, basal-apical calibration, basal-apical length, removal of fingers and hands in the field. The R 4.2.1 software was used, with which a Spearman correlation, analysis of variance (ANOVA) was obtained and for comparison of means, a Kruskal post-hoc was done. Statistical differences were identified (p-value<2.2e-16) and the highest correlations were between age-weight (0.24), age-basal calibration (0.23) and number of hands-weight (0.45). These results indicate that harvest profiling is a practical and agile method that allows quantifying morphological variables of the bunches, making yield projections and adjusting agronomic management depending on the specifications of the marketer.