Landslides, a persistent natural disaster, have historically caused significant loss of life and damage. Recently, human activities have increased the frequency of landslides globally, with Asia accounting for about 75% of these events. To address this, the study proposes a warning system for shallow landslides in Ponmudi, Kerala, India, with potential applicability to similar regions. The research employs cluster and regression analyses to determine rainfall thresholds that trigger landslides in Ponmudi, a highly susceptible area. By analyzing 2-day, 3-day, and 5-day antecedent rainfall data against daily rainfall, critical events likely to cause landslides were identified. The 5-day antecedent rainfall curve, despite some false positives, provided the most reliable threshold equation: y = 55.223−0.6553x. The study also assessed slope stability using the Infinite Slope Analysis Model, implemented through GIS-TISSA. The factor of safety (FS) for different slopes was calculated, with FS values below 1.5 indicating instability. Inputs for this analysis included a Digital Elevation Model (DEM), Normalized Difference Vegetation Index (NDVI) maps, soil maps, and tree root strength data. The research demonstrates that combining rainfall threshold analysis with slope stability assessments offers a cost-effective, comprehensive approach to developing early-warning systems for shallow landslides. This integrated method enhances the ability to predict and mitigate landslides in Ponmudi and similar regions.