Introduction: Craniofacial classification is essential for understanding individual characteristics and has advanced significantly thanks to technological and mathematical developments. However, traditional clinical classifications have limitations that have led to the search for more accurate and objective methods. Data mining and mathematical analysis have been crucial for identifying patterns, predicting growth, and classifying groups with greater precision. These advancements have improved the understanding of the relationships between craniofacial variables and facilitated the identification of specific subgroups. In dentistry, these tools have optimized diagnosis and treatment planning. By analyzing large datasets and using machine learning algorithms, it is possible to develop more accurate and objective classification models, as demonstrated by this study conducted in a Colombian mestizo population, which aimed to develop an automatic classification based on craniofacial characteristics. Method: Hierarchical clustering was used to create natural groupings, proposing and evaluating different numbers of groups or categories. The groups that most coherently described the population were selected, based on the variables with the greatest influence in each group. Conclusion: This study demonstrated the utility of hierarchical clustering for craniofacial cephalometric classification in the era of Big Data. Four biologically distinct categories or groups were established based on craniofacial classifications, providing references for rapid group assignment and facilitating a systematized classification of cephalometric evaluations