Online professional development (OPD) for teachers is an increasingly popular and viable alternative to face-to-face professional development. While OPD can be effective, little is known about OPD's design and implementation features that maximize its impact. Using data from a large-scale OPD initiative, this correlational study (N = 1231) investigates antecedents of self-reported changes in teacher knowledge, classroom practice, and student achievement. Three regression analyses replicate the importance of several factors in effective professional development, or online learning more generally, and also identify additional predictors of OPD's impact(s). The paper also discusses an applied framework for conceptualizing and modeling the effects of OPD's features on its successive outcomes. Implications for the design, implementation and evaluation of OPD, directions for future research, and study limitations are discussed.