This supplement to Hattie et al. explains how the New Zealand Ministry of Education's Iterative Best Evidence Synthesis Programme draws upon meta-analysis within a realist synthesis framework. It discusses the iterative approach to the development and use of research syntheses for educational improvement. The goal of the programme is to make clear evidence of what works, what doesn't work, what makes a bigger difference, and why it matters to education. When the various meta-analytic findings are complemented by cases, underpinned by explanatory theory, and informed a "first do no harm' approach, then the policy and classroom implications of evidence can be made useful to teachers and leaders seeking to invest their time and resources in fruitful ways. The piece underlies the importance of using evidence about change processes as well as the evidence about the focus of change in education as necessary to countering assumptions that educational change will follow from just knowing what makes a bigger difference, as if by magic.