The shape preferred orientation and connectivity of pores in reservoir rocks largely controls fluid migration properties, for example, by defining preferred flow directions. An accurate determination of preferred flow directions, observed as permeability anisotropy, is an integral part of reservoir characterization, due to profound effects on fluid migration. Numerous research fields, including groundwater studies, hydrocarbon exploitation, contamination mitigation, and CO2 sequestration, therefore seek methods to reliably characterize pore fabrics and permeability anisotropy. Many traditional methods face trade-offs between sample size and resolution, and measurements of permeability anisotropy require several oriented cores, where anisotropy may be masked by core-scale heterogeneity, and assumptions on the fabric orientation need to be made when less than six cores are measured. The magnetic pore fabric method has the potential to overcome these difficulties, and has shown promising empirical relationships to both the preferred orientation of pores, and permeability anisotropy. Magnetic pore fabrics are determined by impregnating rock with ferrofluid, and then measuring the anisotropy of magnetic susceptibility. These measurements provide a full 3D average fabric measure from a single core. So far, interpretation was compromised by large variability in the empirical relationships published in different studies. Here, experimental developments, and a conceptual and numerical model are presented that enable more thorough and quantitative interpretation of magnetic pore fabrics.