Housing reconstruction is the top priority after the occurrence of natural hazards. The manner of providing housing not only matters to community recovery but also determines whether reconstruction is sustainable, particularly in rural areas in developing countries. Developing concentrated rural settlements (CRS) is sustainable and resilient in post-disaster reconstruction. However, such development has been reported in few studies before the reconstruction that followed the Sichuan Earthquake on May 12, 2008. CRS have been successfully implemented in several villages in Dujiangyan City, China through a policy of exchanging rural residential land. CRS were delivered in two different approaches, namely, unified planning/self-reconstruction and unified planning/unified–reconstruction. However, these two approaches have not been thoroughly investigated in previous studies. Therefore, the reasons for adopting the two different approaches and which approach is more effective remain unknown. The current study aims to fill in this research gap by comparing the two approaches in the aspects of social–natural basis, trigger event, participants, principles, revenue allocation, process, and implementation outcome involving four study cases in Dujiangyan City. Relevant information was collected through content analysis and interviews. The comparison indicated that neither approach is ideal. Each approach should first satisfy specific requirements to ensure the reconstruction's sustainability. A roadmap to deliver CRS during post-disaster reconstruction has been provided based on the comparison. The findings of this study can facilitate the decision-making process of local governments regarding CRS development during post-disaster reconstruction.