The exceptional versatility of calmodulin (CaM) three-dimensional arrangement is reflected in the growing number of structural models of CaM-protein complexes currently available in the Protein Data Bank (PDB) database, revealing a great diversity of conformations, domain organization, and structural responses to Ca(2+). Understanding CaM binding is complicated by the diversity of target proteins sequences. Data mining of the structures shows that one face of each of the eight CaM helices can contribute to binding, with little overall difference between the Ca(2+) loaded N- and C- lobes, and a clear prevalence of the C-lobe in Ca(2+) free conditions. The structures reveal a remarkable variety of configurations where CaM binds its targets in a preferred orientation that can be reversed, and where CaM rotates upon Ca(2+) binding, suggesting a highly dynamic metastable relation between CaM and its targets. Recent advances in structure-function studies and the discovery of CaM mutations being responsible for human diseases, besides expanding the role of CaM in human pathophysiology, are opening new exciting avenues for the understanding of the how CaM decodes Ca(2+)-dependent and -independent signals.