We analyze theoretically, by means of both computer simulations and laboratory experiments, the limitations of correcting aberrations with ideal customized contact lenses. Four experiments are presented: In the first one, we have analyzed the limitations of a static correction on the dynamic wavefront. In the second one, we studied the rotations of a contact lens on the eye using an optical method. The third one researched the limitations of the wavefront correction, focusing on a group of normal and highly aberrated eyes, when the correction suffers from a permanent rotation or translation. The fourth one estimates, under a simple approximation, the error made when applying on the corneal plane the correction corresponding to the wavefront measured at the entrance-pupil plane. Results show that a static correction of the wavefront leaves a residual aberration of 0.15-0.25 microm for a 5 mm pupil. Rotation of the contact lens (up to +/-4 degrees) diminishes the effectiveness of the correction. Horizontal or vertical translations of 0.5mm could generate a high-order-aberration RMS that is higher than the remaining one after a standard low-order correction. In particular, the group of eyes having normal values of high-order aberrations are more sensitive to translations than the one having higher values. Most of the results could be applied to other methods of aberration correction, such as refractive surgery or correction by means of intraocular lenses.