Interspecific aggression by the noisy miner (Manorina melanocephala), a highly despotic species, is homogenizing woodland avifaunas across eastern Australia. Although a native species, the noisy miner's aggressive exclusion of small birds is a Key Threatening Process under national law. Large-scale removal of noisy miners has been proposed as a management response to this threat following increases in miner presence due to anthropogenic land use practices. We tested this proposal by experimentally removing noisy miners from eucalypt woodland remnants (16-49 ha), assigned randomly as control (n = 12) or treatment (miner removal) sites (n = 12). Standardized bird surveys were conducted before and after removal, and generalized linear mixed models were used to investigate the effect of miner removal on bird assemblage metrics. Despite removing 3552 noisy miners in three sessions of systematic shooting, densities of noisy miners remained similarly high in treatment and control sites, even just 14 days after their removal. However, there was evidence of an increase in richness and abundance of small birds in treatment sites compared to controls-an effect we only expected to see if noisy miner densities were drastically reduced. We suggest that miner removal may have reduced the ability of the recolonizing miners to aggressively exclude small birds, even without substantially reducing miner densities, due to the breakdown of social structures that are central to the species' despotic behaviour. However, this effect on small birds is unlikely to persist in the long term. Synthesis and applications: Despite evidence from other studies that direct removal of noisy miners can result in rapid and sustained conservation benefit for bird communities at small scales, our findings cast doubt on the potential to scale-up this management approach. The circumstances under which direct control of noisy miners can be achieved remain unresolved.