Adding conductive fillers to nonconductive polymers is a common way to make soft conductive materials such as conductive adhesives. An important issue is how to achieve high volume conductivity with acceptable mechanical performance. Two questions pertaining to this issue were studied in this paper. One question was whether the maximum conductivity benefits from larger or smaller particle sizes. The second was what is the maximum achievable conductivity. One incentive for this work is the recent availability of nanomaterials that provide opportunities to make conductive composites using much smaller particles than in the past. We found that the conductivity of platinum, carbon black, and silver particles in their polyurethane composites did not vary greatly with particle size (from micrometer to nanometer range). What was unexpected was that in all the composite examples, the highest conductivity achieved was only on the order of 1% of that of the pure bulk conductive materials. Further experiments to emulate these conductive composites with platinum, carbon black, copper, and nickel particles without polymer matrix showed similar results, indicating the issue is not simply dispersion homogeneity, nano versus macro particles, particle connectivity/percolation, or the presence of the matrix materials. We interpret this to mean that the composite systems are intrinsically limited by the contact between filler particles.