Human genetic variation is both highly complex and deeply hierarchical, making its characterization challenging. Clustering techniques play a central role in how we interpret genetic variation and reconstruct population history, being widely used in genome-wide association studies. In this context, equipping clustering methods (which are inherently exploratory) with inferential tools may help strengthen the reliability of genetic data analyses and, more importantly, clarify the non-trivial interpretation of genetic clusters. In this talk, we introduce the theory of post-clustering inference and assess its suitability when applied to genotype datasets. In particular, we show how these techniques can be used to endow a dendrogram with p-values and to define a more appropriate branch-cutting strategy for deriving a partition. This results in a p-value-based clustering procedure that performs well in recovering the population structure of several human populations.