A recent issue of the Australian Journal of Language and Literacy included an article reporting on a systematic narrative review of the research literature that indicated that there was insufficient evidence to conclude whether genre theory and systemic functional linguistics either 'worked' or 'did not work'. The criteria used to evaluate these studies excluded any study that did not conform to the 'gold standard' associated with experimental research such as randomised controlled trials. In response to this provocative finding, a group of SFL researchers decided to examine just what counts as evidence of quality literacy research these days. In this paper, we question the overreliance on experimental research at the expense of other methods. We illustrate this with a sample of notable studies that do not meet experimental criteria, but which nevertheless have made a significant contribution to school literacy outcomes in Australia and elsewhere.