Persistent disparities in access to healthcare, infrastructure, education, and social services across marginalized regions have prompted the need for more innovative tools in identifying and prioritizing underserved communities. Conventional data collection methods—censuses, household surveys, and administrative records—often fail to capture rapidly changing socio-spatial dynamics or adequately represent remote, informal, or high-mobility populations. In this context, remotely sensed data from satellite and aerial imagery presents a powerful alternative, offering scalable, repeatable, and spatially comprehensive insights. This study presents a project-based approach to identifying underserved populations by integrating satellite-derived indicators such as night-time light intensity, land use classification, and impervious surface distribution with ancillary geospatial datasets. Through an iterative process involving data preprocessing, feature extraction, and supervised classification, the methodology allows for the visualization of service deprivation zones that align with qualitative ground-truthing efforts. Case examples include peri-urban expansion zones, informal settlements, and rural agricultural corridors with low infrastructure density and poor connectivity. The approach demonstrates how open-access earth observation data can be harnessed through low-cost, replicable workflows to support evidence-based targeting of development interventions. Importantly, the project-based model emphasizes skill-building in GIS and remote sensing techniques among local planning teams, making it suitable for integration into educational and capacity-building programs. The results suggest that even modest analytical infrastructures can yield actionable insights, contributing to inclusive policy design, humanitarian programming, and equitable urban development strategies.