Experimental infections of insects with arboviruses are performed to achieve a variety of objectives but principally to draw inferences about the potential role of field populations in transmission or to explore the molecular basis of vectorpathogen interactions. The design of such studies determines both their reproducibility and the extent to which their results can be extrapolated to natural environments, and is constrained by the resources available. We discuss recent findings regarding the effects of nutrition, the microbiome, co-infecting agents and feeding methods on the outcome of such experiments, and identify resource-efficient ways to increase their relevance and reproducibility, including the development of community standards for reporting such studies and better standards for cell line and colony authentication.