Microfacet-based material appearance models are commonly considered as a physical plausible representation of matter-light interaction. With such models, the microgeometry of a surface element is defined by a statistical distribution of microfacets. The mathematical formulation ensures physical plausibility, such as energy conservation and reciprocity. Many authors have addressed microfacet bidirectional scattering distribution function (BSDF) representations, with various normal distribution functions (NDFs) and their relationship with shadowing and masking, or the effects due to multiple light scattering on the microgeometry. However, an extensive study on how an actual microgeometry drives material appearance still is missing. This question is a key issue for inverse design and manufacturing. This paper contributes to filling this gap by proposing a complete pipeline composed of a microgeometry generation process and numerical lighting simulation. From any input NDF, our method generates a controlled and structured microgeometry, integrated within numerical light scattering simulation. Reflected light is gathered using a virtual goniophotometer. From a given set of parameters, we use our pipeline to study the impact of microgeometry structures on material light scattering in the case of rough surfaces. The obtained results are discussed and compared with already existing approaches when they exist in the pipeline.