In optimizing a design of an integrated circuit, an iteration of a logic optimization process is performed that at least partially optimizes a circuit design such that there is slack remaining in one or more combinational logic paths in the circuit design following the iteration. A clock latency scheduling process is performed that respectively distributes the remaining slack of one or more respective combinational logic paths in the circuit design across respective registers in the circuit design. Another iteration of the logic optimization process is performed that uses at least a portion of the distributed slack to further optimize the circuit design.