Traditionally, structure-based drug design has been predicated on the idea of the lock-and-key hypothesis, i.e., the ideal drug should have a structure that complements the target site structurally and energetically. The implementation of this idea has lead to the development of drug molecules that are conformationally constrained and pre-shaped to the geometry of the selected target. The main drawback of this strategy is that conformationally constrained molecules cannot accommodate to variability in the target and, therefore, lose significant binding affinity even in the presence of small changes in the target site. There are three common situations that lead to binding site heterogeneity: (1) genetic diversity; (2) drug resistant mutations; and (3) binding site dynamics. The development of drugs that effectively deal with target heterogeneity requires the introduction of certain degree of flexibility. However, flexibility cannot be introduced indiscriminately because it would lead to a loss of binding affinity and specificity. Recently, structure-based thermodynamic strategies aimed at developing adaptative ligands that target heterogeneous sites have been proposed. In this article, these strategies are discussed within the context of the development of second generation HIV-1 protease inhibitors.