Outcome studies in dialysis have generally failed to show an impact of changes in membrane flux or biocompatibility, and only dose increases up to a certain level have been shown to improve survival significantly. However, to see an effect of a potentially improved dialysis treatment, all available factors that make dialysis more physiological may need to be combined. A membrane that mimics the glomerular basement membrane in being hemocompatible, having a high hydraulic permeability and a generous sieving, yet not leading to albumin loss could be used. The dialysis fluid composition could be individualized, and the quality and volume appropriate for the selected application. The new system of online-prepared ultrapure dialysis fluid and sterile infusion solutions, as integral parts of the treatment, are cost effective and labor saving as well as biocompatible. Ideally, we should select a blood purification method that covers the same range of solutes as the kidney. Convection is equally effective for all solutes that can pass through the membrane, and the corresponding renal therapy is hemofiltration. For enhanced small solute removal, convection can be combined with diffusion as in hemodiafiltration, which has the potential to achieve the largest solute removal over a wide molecular weight spectrum among all forms of dialysis. Finally, the dialysis treatment should be applied as often as is practically possible-preferably daily-in order to reduce the peaks caused by uremic toxins, the exposure to acidosis and alkalosis, and the burden on the cardiovascular system by overhydration. While the designed therapy is already technically feasible today, a positive result from outcome studies will be needed to bring about the political and economic decisions required to change conventional dialysis into a treatment approaching true renal blood purification.