Research on granulation processes has concentrated on the use of mechanical mixers. Understanding of the mechanisms by which granules are formed interact with each other and change in size has increased greatly. We now appreciate why products frequently have a bimodal distribution, wide size range and non-uniform binder content. The effects of changing binder viscosity and size of the constituent particles are also partially understood. However, much remains to be done. Three areas of research are suggested that may repay intensive investigation. The first challenge is to improve knowledge of the strength of wet assemblies. This is fundamental to granule deformation and coalescence processes and yet is not well understood. Another challenge is to develop better models for granule coalescence. Although there have been significant advances on understanding of the processes of granule adhesion and coalescence, more needs to be done. The third challenge is to learn how to design mixers that inherently give a better control of granule size. This requires an understanding of the motion of material within granulators and how the granulator interacts with the material being granulated. (C) 2004 Elsevier B.V. All rights reserved.