Polymer membrane-based desalination technologies (e.g., reverse osmosis (RO) and nanofiltration (NF)) have been extensively developed since the 1960s and are well-established processes. The separation performance of polymeric membranes for desalination is usually described in terms of water flux (or permeance) and salt rejection. Based on a survey of available data, water permeance and NaCl rejection are often inversely correlated, and there may be an upper bound, similar to that observed in gas separation membranes, beyond which there are very few data points. However, water permeance and salt rejection are not intrinsic material properties since they are influenced by sample size (i.e., membrane thickness in the case of permeance) and measurement variables (e.g., pressure and salt concentration in the case of salt rejection). Use of water permeability, rather than water flux or permeance, and water/salt permeability selectivity, rather than rejection, in a tradeoff analysis provides a clearer comparison of properties that depend only on the fundamental transport characteristics of the materials under study. When water and salt transport data are presented on a log-log plot of water permeability versus water/NaCl permeability selectivity, a clear tradeoff relation and upper bound emerges. Further analysis suggests that both water/NaCl solubility and diffusivity selectivity contribute to the high water/NaCl permeability selectivity, but diffusivity selectivity is the dominant factor. Both solubility selectivity and diffusivity selectivity exhibit tradeoff and upper bound features when plotted as a function of water solubility and water diffusivity, respectively.