If we are to accelerate the recolonisation of restored areas by slow-recolonising species, we must provide suitable microhabitats at appropriate densities. Previous research in south-western Australia has shown that Napoleon’s skink (Egernia napoleonis) rarely recolonises restored areas. We trapped Napoleon’s skink in restoration and unmined forest to confirm the species was late successional. We also radio-tracked six skinks in unmined forest, to determine types and characteristics of used microhabitats, and estimated home ranges to determine required microhabitat densities, with the aim of accelerating skink recolonisation of restored areas. All tracked skinks used logs and hollow-bearing trees. Used logs were larger, and used trees were larger and taller than random samples, probably because large logs and trees were more likely to contain cracks and hollows that provide a refuge from predators. Extrapolations from home-range estimates indicated that a minimum of four logs ha–1 are required in restored areas to facilitate recolonisation by skinks, with skink densities likely to increase with log densities. Our study demonstrated that not all fauna will naturally recolonise restored areas and management of these areas is required to provide suitable habitat for late-successional species. Our approach could potentially be applied to other ecosystems or species.