Existing masonry structures have demonstrated over time to be weak under horizontal loads, especially under seismic actions. For this reason, in recent years the importance of structural interventions to repair and strengthen this kind of structures has increased considerably. Among the several approaches available to retrofit masonry buildings, the traditional steel reinforced plaster (SRP) is largely adopted in the daily practice to reinforce buildings without special historical and cultural value, even though the scientific literature and the standards do not provide a proper treatment of the method nor give guideline for design. This paper presents the results of an experimental program focused on the evaluation of the performance of masonry specimens retrofitted with SRP under cyclic loads to support the development of a design concept. A total of 14 specimens (4 unreinforced + 10 retrofitted) were tested under displacement controlled diagonal compression. Several design parameters were investigated: the thicknesses of masonry, the mechanical properties and thickness of the plaster cover, the number of connectors. The results showed an improved performance for the retrofitted specimens with respect to the unreinforced ones, in terms of both strength and deformation capacity. In particular, a significant ductility was reached by the specimens with higher number of anchors and higher thickness of the plaster layers, while a higher peak load was associated in some cases to thicker plasters. Finally, the peak load seemed to be not affected by the type of plaster mortar and the number of anchors.