Current regional landslide Early Warning Systems (LEWS) are typically based on rainfall data and rarely use soil wetness information in spite of the direct relationship between soil wetness and landslide triggering. This is partly due to the limited availability of soil moisture measurement networks and the lack of experience in interpreting such data. Here we show how soil wetness measurements could be exploited for LEWS in a region (or country) with variable topography and soil composition. In particular, we highlight the relevance of (steep) topography for the soil wetness observations and its significance for the interpretation in regard to landslide criticality. To this end, a field study has been set up and is currently running in a landslide-susceptible region of central Switzerland (Emmental) where different soil wetness measurement methods are compared on a flat and a steep meadow site. First results suggest that in spite of noticeable topographical effects on soil drying and wetting the informative value of these data for LEWS is not very different for sloped and flat locations. Based on these findings and a comprehensive analysis of the co-occurrence between soil wetness indicators and shallow landslides across Switzerland we attribute soil wetness measurements a significant value for landslide early warning and encourage a wider incorporation of such data in existing warning systems.