Current guidelines suggest that stabilized rammed earth materials be compacted at their optimum water content to achieve their maximum strength. Although this is true for traditional rammed earth, there is no evidence that this procedure should also be used for cement-stabilized rammed earth. Furthermore, the water content used at compaction is usually difficult to control on a construction site, so that material might be compacted at water contents other than the optimum. In this paper, a novel experimental programme is presented in which the effect of compaction water content on the unconfined compressive strength of crushed limestone stabilized to 5% Portland cement content is investigated for a range of curing periods. Freeze drying of specimens was used to arrest cement hydration to determine the evolution of hydrated cement content. Scanning electron microscope (SEM) analysis was used to identify differences between the final material microstructures. Results are discussed demonstrating the intimate link between the amount of hydrated cement, material microstructure, and compressive strength.