Methods based on Computational Fluid Dynamics (CFD) are a promising tool to gain insight into poorly accessible and swift solid formation processes such as granulation or precipitation. Unfortunately, CFD modeling of transient particle size distributions in such reactors requires enormous computational effort because complex and multi-scale interactions have to be considered. Hence, a new coarse graining approach is presented with which these complex interactions can be handled with reasonable computational effort and which enables to extract short cut methods from complex CFD simulations. The method is exemplified for precipitation crystallization of nanoscaled solid particles from the liquid phase. This process is governed by fast primary processes, such as super saturation build-up, nucleation and growth. Experimental access to internal parameters is very difficult or even impossible. The new methodology which we call "Spatially and Temporally Averaged Reduced Numeric Measurement" (STAR NM) is based on a reasonable averaging and correlation of process dominating state variables or rates which are consigned to a fast 1D population balance solver. Precipitation of the sparingly soluble barium sulfate in water is used as a model process. A confined impinging jet mixer is used as a benchmark apparatus which allows for the adjustment of highly reproducible mixing conditions. Experimental results are compared to those gained with the new STAR NM approach. (C) 2016 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.