In pole assignment for multi-input systems the constant gain feedback control to achieve desired closed-loop poles is not unique. Neither is the feedback control for single-input systems of which not all poles are assigned. It is proposed to use this design freedom to minimize a given performance function, thus combining the pole placement and optimal control design techniques, From a different point of view, the method augments the pole placement technique by a method of implicit optimal zero placement. The numerical technique for solving the single-input case is detailed in this paper.