Ecological fire management in Australia is often built on an assumption that meeting the needs of plant species will automatically meet the needs of animal species. However, the scarcity of ecological data on the needs of fauna in relation to fire undermines the confidence managers should place in current popular frameworks for planning ecological burning. Such frameworks are built almost entirely around the goal of maintaining plant community diversity. They provide little guidance to managers regarding the characteristics of desirable ‘mosaics’ (e.g. patch size, connectivity or composition of age-since-burnt classes) or the timing of fires in relation to faunal population trends linked to other cycles (e.g. El Niño events). Claims by agencies of adopting an adaptive management approach (‘learning by doing’) to cope with a dearth of knowledge are credible only if monitoring and evaluation are carried out and future actions are modified in light of new evidence. Much monitoring of fauna is of such a small scale and short duration that the statistical likelihood of detecting a positive or negative effect of the management regime is minute. Such shortcomings will only be overcome through broad-scale and/or long-term studies of fauna. The funding for such research is unlikely to be forthcoming if fire ecologists and land managers convey the impression that the current data are adequate for the implementation of the current planning frameworks.