When trying to optimize hemodialysis adequacy, it can be questioned whether one should focus on the dialyzer or on the patient. Another crucial question is whether the currently applied dialysis adequacy parameter, Kt/Vurea, is a reliable marker. For the small and water-soluble solutes, recent advances in convective strategies and/or new dialyzer designs do not add much removal capacity. Depending on their specific kinetics, generally quite different from those of urea, small solute removal benefits from longer or more frequent dialysis. Clearance of beta-2-microglobulin (β2M), a marker of middle molecule removal pattern, is improved with dialysis using more open and permselective membranes, as well as by using high convective volume strategies. Furthermore, longer and more frequent dialyses have highly favorable removal characteristics because they facilitate the retarded transport between plasmatic and extraplasmatic compartments over which these molecules are distributed. As β2M may not be representative of other middle molecules, future kinetic analyses of alternative middle molecules will be of the utmost interest. Protein-bound solute clearance is improved by convective techniques, but not by more open dialyzer pores. Knowledge of their kinetics should be helpful in interpreting the observation that frequent (but not longer) dialysis enhances protein-bound solute removal. Hence, further technical improvements in dialyzers will have only a minor impact on dialysis adequacy, as retarded solute movement in the patient plays a decisive role. As urea kinetics is not representative of the kinetics of protein-bound compounds, middle molecules, nor even of other small and water-soluble solutes, it becomes self-evident that urea clearance is a poor predictor of many aspects of dialysis adequacy.