Adhesion between metal surfaces and soil can have significant effects on soil-tool interaction. The cone pull-out test (CPT) and piston pull-out test (PPT) for adhesion assessment were used in this study on smectite, illite and kaolinite clays at different water contents. In the CPT, the tangential adhesion stress (TAS) and cone adhesion (CA) were measured. In the PPT, the normal adhesion stress (NAS) and piston adhesion (PA) were measured. The results showed that the NAS and TAS depended on the type of clay. The ratios of NAS to TAS for smectite and kaolinite had the largest and smallest values, respectively. For all three types, NAS and TAS decreased as the water content (W) decreased. An increase in the thickness of the water film at the soil-metal contact surface reduced the stress required to separate the metal from the soil. The relationship between CA-W and PA-W for the three clay types differed. For smectite, a change in water content had no effect on CA or PA. For illite, an increase in water content decreased the CA and PA. For kaolinite, a bell-shaped curve formed such that, as water content increased, where WLL, CA and PA decreased. For all three clay soil types, NAS and PA were higher than TAS and CA at a specific water content. This demonstrated the effect of the metal-to-soil contact geometry on the adhesion parameters. The NAS and TAS values were classified in terms of clogging potential. Clogging potential is the adhesion of soil to the cutting tools or conveying system. The selection of the use of the CPT or PPT for adhesion assessment depended on the phenomenon being studied. • The metal-to-soil contact geometry does affect on adhesion parameters. • The normal adhesion stress (NAS) was higher than the tangent adhesion stress (NAS). • The piston adhesion (PA) was higher than the cone adhesion (CA). • The NAS and TAS decreased as the water content decreased. • There is linear relationship between NAS and TAS.