Commuting flows refer to the regular movement of people from their homes to workplaces. The spatial heterogeneity of commuting flows indicates the uneven spatial distribution of places of residence, employment, or job–residence connections. Understanding the characteristics of spatial heterogeneity in commuting flows is vital for effective transportation planning. However, limited by the scarcity of flow data, traditional research on commuting flows predominantly focuses on the spatial distribution of employment, residence, and job–residence connections individually, which fails to unveil the spatial heterogeneity of commuting flows. In this study, we examined the spatial heterogeneity of commuting flows in Beijing using mobile phone data. We analyzed the degree of heterogeneity and the aggregation scale of commuting flows. The results showed that (1) the degree of spatial heterogeneity varies between different regional pairs, and (2) the aggregation scale of commuting flows varies with distance; moreover, the longer the distance, the larger the aggregation scale of commuting flow distribution, and dominant heterogeneous clusters expand from the center to the periphery. These findings enhance the strategic planning of public transit. In less densely populated areas such as those outside the Fifth Ring Road, efficiency can be boosted by new bus routes. In contrast, denser areas such as those within the Third Ring Road, may benefit from an integrated approach: establishing new rail systems alongside expanded bus services to optimize commuting.