Regulation of molecular transport via intercellular channels called plasmodesmata (PDs) is important for both, the coordination of developmental and environmental responses among neighbouring cells and the isolation of cell domains to execute specific developmental or stress-induced programs. PD transport capacity (i.e. effective symplastic permeability) has been determined experimentally, at a tissue level, by assessing the mobility of different fluorescent molecules, or predicted from PD ultrastructural features using electron-microscopy. Values obtained from these approaches are often very different. Here, we build a theoretical bridge between the two experimental approaches by calculating the effective symplastic permeability from a geometrical description of individual PDs, considering the flow towards them and including the impact of PD clustering into pit fields. Our open source multi-level model allows us to link measured permeabilities with realistic PD dimensions, predict how typical PD features affect transport properties and add a functional interpretation to recent experimental observations.