The advancement of 3D printing and scanning technology enables the digitalization and customization of foot orthosis with better accuracy. However, customized insoles require rectification to direct control and/or correct foot deformity, particularly flatfoot. In this exploratory study, we aimed at two design rectification features (arch stiffness and arch height) using three sets of customized 3D-printed arch support insoles (R+U+, R+U−, and R−U+). The arch support stiffness could be with or without reinforcement (R+/−) and the arch height may or may not have an additional elevation, undercutting (U+/−), which were compared to the control (no insole). Ten collegiate participants (four males and six females) with flexible flatfoot were recruited for gait analysis on foot kinematics, vertical ground reaction force, and plantar pressure parameters. A randomized crossover trial was conducted on the four conditions and analyzed using the Friedman test with pairwise Wilcoxon signed-rank test. Compared to the control, there were significant increases in peak ankle dorsiflexion and peak pressure at the medial midfoot region, accompanied by a significant reduction in peak pressure at the hindfoot region for the insole conditions. In addition, the insoles tended to control hindfoot eversion and forefoot abduction though the effects were not significant. An insole with stronger support features (R+U+) did not necessarily produce more favorable outcomes, probably due to over-cutting or impingement. The outcome of this study provides additional data to assist the design rectification process. Future studies should consider a larger sample size with stratified flatfoot features and covariating ankle flexibility while incorporating more design features, particularly medial insole postings.