The life and death of an organism often depends on its ability to perform well at some ecologically relevant task. Yet despite this significance we have little idea how well species are optimised for competing locomotor tasks. Most scientists generally accept that the ability for natural systems to become optimised for a specific task is limited by structural, historic or functional constraints. Climbing lizards provide a good example of constraint where climbing ability requires the optimization of conflicting tasks such as speed, stability, or efficiency. Here we reconstruct multiple performance landscapes of climbing locomotion using a 10-DOF robot based upon the lizard bauplan, including an actuated spine, shoulders, and feet, the latter which interlock with the surface via claws. This design allows us to independently vary speed, foot angles, and range of motion, while simultaneously collecting data on climbed distance, stability and efficiency. We first demonstrate a trade-off between speed and stability with high speeds resulting in decreased stability and low speeds an increased cost of transport. By varying foot orientation of fore and hindfeet independently, we found geckos converge on a narrow optimum for both speed and stability, but avoid a secondary wider optimum highlighting a possible constraint. Modifying the spine and limb range of movement revealed a gradient in performance. Evolutionary modifications in movement among extant species appear to follow this gradient towards areas which promote speed and efficiency. This approach can give us a better understanding about locomotor optimization, and provide inspiration for industrial and search-and-rescue robots.Significance StatementClimbing requires the optimization of conflicting tasks such as speed, stability, or efficiency, but understanding the relative importance of these competing performance traits is difficult.We used a highly modular bio-inspired climbing robot to reconstruct performance landscapes for climbing lizards. We then compared the performance of extant species onto these and show strong congruence with lizard phenotypes and robotic optima.Using this method we can show why certain phenotypes are not present among extant species, illustrating why these would be potentially mal-adaptive.These principles may be useful to compare with relative rates of evolution along differing evolutionary histories. It also highlights the importance of biological inspiration towards the optimization of industrial climbing robots, which like lizards, must negotiate complex environments.