Variety mixtures can provide a range of benefits for both the crop and the environment. Their utility for the suppression of pathogens, especially in small grain crops, is well established and has seen some remarkable successes. However, despite decades of academic interest in the topic, commercial efforts to develop, release and promote variety mixtures remain peripheral to normal breeding activities. Here we argue that this is because simple but general design principles that allow for the optimization of multiple mixture benefits are currently lacking. We therefore review the practical and conceptual challenges inherent in the development of variety mixtures, and discuss common approaches to overcome these. We further consider three domains in which they might be particularly beneficial: pathogen resistance, yield stability and yield enhancement. We demonstrate that combining evolutionary and ecological concepts with data typically available from breeding and variety testing programmes could make mixture development easier and more economic. Identifying synergies between the breeding for monocultures and mixtures may even be key to the widespread adoption of mixtures-to the profit of breeders, farmers and society as a whole.