Population analysis is crucial for ensuring that empirical software engineering (ESE) research is representative and its findings are valid. Yet, there is a persistent gap between sampling processes and the holistic examination of populations, which this position paper addresses. We explore the challenges ranging from analysing populations of individual software engineers to organizations and projects. We discuss the interplay between generalizability and transferability and advocate for appropriate population frames. We also presents a compelling case for improved population analysis aiming to enhance the empirical rigor and external validity of ESE research.