Purpose: To refine the refractive outcome of the second eye after cataract surgery by deriving adjustment coefficients for intraocular lens (IOL) selection based on the prediction error (PE) of the first eye. Setting: University Hospital Southampton, Southampton, England, and the Lions Eye Institute, Perth, Australia. Design: Retrospective study of two heterogeneous datasets. Methods: One hundred thirty-nine patients who underwent delayed sequential bilateral cataract surgery in Australia were retrospectively analyzed. The PE was determined by comparing postoperative subjective refraction with the predicted postoperative refraction (PPOR) calculated by the Barrett Universal II, Hoffer Q, Holladay I, and SRK/T formulas. Adjustment coefficients were derived for each formula and applied to the second eye's IOL calculation. Separately, patient-specific optimized IOL constants were derived from the first-eye PE and applied to the second-eye calculation. The same adjustments were applied to a dataset of 605 patients in the United Kingdom to test the validity of the Australian results. Results: The study comprised data on 139 patients in Australia and 605 patients in the U.K. The Australian-derived adjustment coefficients based on PE ranged from 0.30 to 0.56 (Barrett Universal II 0.30; Hoffer Q 0.56; Holladay I 0.53; SRK/T 0.48). Applying these to the U.K. dataset led to the percentage of patients within 0.50 diopters of PPOR with their second eye improving from 70.74%, 65.29%, 69.09%, and 67.77%, with the Barrett Universal II, Hoffer Q, Holladay I, and SRK/T, respectively, to 72.73%, 68.76%, 71.57%, and 72.56%. Using patient-specific optimized IOL constants derived from the first eye had similar efficacy to formula-specific adjustment. Conclusion: Second-eye refinement via either formula-specific PPOR adjustment or patient-specific IOL constant adjustment improved the percentage of patients achieving the refractive target with their second eye.