Microbially induced reductions in the saturated hydraulic conductivity, Ks, of natural porous media, conventionally called bioclogging, occurs frequently in natural and engineered subsurface systems. Bioclogging can affect artificial groundwater recharge, in situ bioremediation of contaminated aquifers, or permeable reactive barriers. In this study, we designed a series of percolation experiments to simulate the growth and metabolism of bacteria in sand columns. The experimental results showed that the bacterial cell amount gradually increased to a maximum of 8.91 log10 CFU/g sand at 144 h during the bioclogging process, followed by a decrease to 7.89 log10 CFU/g sand until 336 h. The same variation pattern was found for the concentration of tightly bound extracellular polymeric substances (TB-EPS), which had a peak value of 220.76 μg/g sand at 144 h. In the same experiments, the concentration of loosely bound extracellular polymeric substances (LB-EPS) increased sharply from 54.45 to 575.57 μg/g sand in 192 h, followed by a slight decline to 505.04 μg/g sand. The increase of the bacterial cell amount along with the other two concentrations could reduce the Ks of porous media, but their relative contributions varied to a large degree during different percolation stages. At the beginning of the tests (e.g., 48 h before), bacterial cells were likely responsible for the Ks reduction of porous media because no increase was found for the other two concentrations. With the accumulation of cells and EPS production from 48 to 144 h, both were important for the reduction of Ks. However, in the late period of percolation tests from 144 to 192 h, LB-EPS was probably responsible for the further reduction of Ks, as the bacterial cell amount and TB-EPS concentration decreased. Quantitative contributions of bacterial cell amount and the two types of extracellular polymers to Ks reductions were also evaluated.