Twin-screw granulation is a promising continuous alternative for traditional batch high shear wet granulation (HSWG). The extent of HSWG in a twin screw granulator (TSG) is greatly governed by the residence time of the granulation materials in the TSG and degree of mixing. In order to determine the residence time distribution (RTD) and mixing in TSG, mostly visual observation and particle tracking methods are used, which are either inaccurate and difficult for short RTD, or provide an RTD only for a finite number of preferential tracer paths. In this study, near infrared chemical imaging, which is more accurate and provides a complete RTD, was used. The impact of changes in material throughput (10–17 kg/h), screw speed (500–900 rpm), number of kneading discs (2–12) and stagger angle (30–90°) on the RTD and axial mixing of the material was characterised. The experimental RTD curves were used to calculate the mean residence time, mean centred variance and the Péclet number to determine the axial mixing and predominance of convective over dispersive transport. The results showed that screw speed is the most influential parameter in terms of RTD and axial mixing in the TSG and established a significant interaction between screw design parameters (number and stagger angle of kneading discs) and the process parameters (material throughput and number of kneading discs). The results of the study will allow the development and validation of a transport model capable of predicting the RTD and macro-mixing in the TSG. These can later be coupled with a population balance model in order to predict granulation yields in a TSG more accurately.