AbstractExperimental studies assessing the impact of demographic changes on aggression and inter‐group competitive outcomes in communities of social species are rare. This gap in our knowledge is important, not only because social species are foundational elements of many terrestrial ecosystems, but because interference competition among social groups often involves decision‐like processes influenced by demographic and environmental contexts. In East Africa, the symbiotic ant Crematogaster mimosae is a co‐dominant competitor that engages in high‐mortality, intra‐ and interspecific battles for sole possession of host trees. We manipulated worker density on C. mimosae Acacia host trees, and the colony's opportunity to expand onto neighboring trees to identify how these factors influenced colony‐level aggression, expansion success, and longer‐term survivorship. Worker density on focal trees was increased through translocation of domatia‐bearing branches, and was decreased using partial tree fumigations. We examined impacts of density manipulations on aggression and immediate expansion success under two different risk scenarios. We tied focal trees to either an experimentally emptied‐tree (low‐risk treatment), or to a C. nigriceps‐occupied tree (high‐risk treatment). Expansion success onto emptied neighbor trees was 100% for controls and increased‐density colonies, but only 50% for decreased‐density colonies, despite the fact that host trees are a limiting resource in this system. Success expanding onto trees occupied by a heterospecific competitor reached 36%, 40%, and 79% in decreased, control, and increased‐density trees, respectively. Our results show that changes in worker density due to disturbances or inter‐group battles have the potential to disrupt competitive hierarchies. Worker density manipulations also affected longer‐term colony persistence. Behavioral and genetic data revealed that 12 months after expansions 100% of the decreased‐density colonies, and 25% of control and increased‐density colonies, had been supplanted by neighboring opportunistic conspecifics. Intraspecific aggression may have lower costs in C. mimosae because aggressive colonies can incorporate workers or queens from defeated competitors. The unexpectedly high frequency of conflicts between conspecific C. mimosae, in combination with behaviors decreasing the cost of intraspecific competition relative to interspecific conflict, may create opportunities for the survival of subordinate competitors in this ant–plant system.