In this work, we propose a methodology to couple rainfall thresholds and susceptibility maps in the frame-work of regional scale landslide warning systems. While statistical rainfall thresholds are used to accomplish a temporal forecasting with very coarse spatial resolution, landslide susceptibility maps provide a static spatial information about the probability of landslide occurrence. If the susceptibility map is subdivided in a number of susceptibility classes that equals the number of possible alert levels featured in the threshold system, it is possible to establish a correspondence where the higher the alert level, the lower the degree of susceptibility that could be actually interested by landslides. Using past rainfall and landslide data, we tested this approach in two test sites, corresponding to the Emilia Romagna region ad to an experimental area within Tuscany region, Italy. Results show that the proposed approach would have contributed to define a more accurate location for up to 93% of the landslides correctly forecasted by the regional warning system. Civil protection agencies could use the proposed approach during the alert phases to better define the areas that are possibly going to be affected by landslides. The proposed approach is easy to implement and could be applied to any case of study in which rainfall thresholds and susceptibility maps have been previously defined.