Messenger RNA (mRNA) is a promising tool for biotherapeutics, and self-amplifying mRNA (saRNA) is particularly advantageous as it results in abundant protein expression and production is easily scalable. While mRNA therapeutics have been shown to be highly effective in small animals, the outcomes do not scale linearly when these formulations are translated to dose-escalation studies in humans. Here, we utilize a Design of Experiments (DoE) approach to optimize the formulation of saRNA lipid nanoparticles in human skin explants. We first observed that luciferase expression from saRNA peaked after 11 days in human skin. Using DoE inputs of complexing lipid identity, lipid nanoparticle dose, lipid concentration, particle concentration, and ratio of zwitterionic to cationic lipids, we optimized the saRNA-induced luciferase expression in skin explants. Lipid identity and lipid concentration were found to be significant parameters in the DoE model, and the optimized formulation resulted in ~7-fold increase in luciferase expression relative to initial DOTAP formulation. Using flow cytometry, we observed that optimized formulations delivered the saRNA to ~2% of the resident cells in the human skin explants. Although immune cells make up only 7% of the total population of cells in skin, immune cells were found to express ~50% of the RNA. This study demonstrates the powerful combination of using a DoE approach paired with clinically relevant human skin explants to optimize nucleic acid formulations. We expect that this system will be useful for optimizing both formulation and molecular designs of clinically translational nucleic acid vaccines and therapeutics.