The two features in this issue tell of recovery of local ecosystems after thinking ‘outside the square’. The first is the case of Hinewai Reserve in New Zealand, owned and managed by the Maurice White Native Forest Trust, where the paper's author Hugh Wilson recognised that the colonising shrub Gorse (Ulex europaeus) facilitated rather than inhibited recovery by the site's prior species. The second is the case of East Trinity wetland, near Cairns Queensland, Australia where the Queensland Government has used lime-assisted tidal inundation to counter degradation severely degraded coastal acid sulphate soil created by ill-advised agriculture. In both cases, outstanding natural recovery has occurred to date and is continuing. There are many take home messages from these case studies – not only the obvious messages regarding potential to harness the resilience of local species to periods of stress and stress release but also messages regarding what is the appropriate level of intervention needed to achieve recovery. This emphasises that all restoration interventions are, in essence, methods of compensating for damage to ensure that natural recovery – the purpose and end result of all restoration – can recommence. Those lessons are obvious in projects where natural recovery is clearly the primary mechanism of reinstating biota; but they are less obvious in projects where all habitat features and biota have been depleted and timely recovery requires rebuilding habitats and reintroducing biota. In mine site restoration, for example, soils, habitat structure and biota usually need to be completely reintroduced; yet it is still rewarding to model interventions on mimicking natural conditions and processes; whether it include such things as preparations for germination and plant development (see Mayence and colleagues this issue) or importing woody debris to foster invertebrate recovery (see Lythe and colleagues). Ecological restoration and management requires us to avoid simplistic dogmas where they are not helpful. Not only does it behove us to question assumptions such as that only native plants can play a role in native recovery – but also assumptions such as that all natives necessarily behave in a manner conducive to the persistence of the broader native ecosystem. Barrett and colleagues, for example, find somewhat damaging levels of browsing by native fauna on an endangered ecological community in the Stirling Range in Western Australia and suggest potential management options. Thinking outside the square is also required to manage for unexpected threatened species in recovering mine sites (Craig and colleagues); differing fire regimes where threatened species share a habitat (Hunter); finding methods for detecting whether management is achieving appropriate conditions for target native fauna (Schultz and colleagues); and considering new funding models for ecological research (Dovers and colleagues). The paper in this issue reviewing progress with implementation of the Wunambal Gaambera People's Healthy Country Plan also signals the maturation of efforts to do things differently, adapting Conservation Action Planning to Indigenous land and sea management in ways that lead to specific conservation and culturally beneficial outcomes (see Austin and colleagues this issue). Just like in the case of Hinewai Reserve in New Zealand, Australian cases of private conservation management planning can be flexible enough to build capacity and ensure the planning and implementation is socially responsive and can be sustained over time.