Summary Despite well-established knowledge about the relationships between the electrical conductivity or relative permittivity of rocks, ambiguity remains as to how these relationships and knowledge can be transferred to soils. In general, there are three problems that hinder the robust application of available pedophysical models. First, these models often rely on soil properties and attributes that cannot be directly quantified in the laboratory or in the field, such as pore connectivity and depolarisation factor, which reduces their applicability. Secondly, the introduction of tuning parameters (such as the cementation exponent and Roth's exponent) tends to reduce the theoretical significance of the physical process in question. Finally, oversimplifying pedophysical relationships by considering only one soil attribute (such as moisture content) effectively ignores the effect of other relevant properties (clay content or bulk density). Here we present a theoretical approach that shows how to sort out this issues while both fitting parameters and anisotropy indicators can be derived from soil properties that are easy to measure in the laboratory.