Recent studies in mice have shown that pancreatic β-cells have a significant potential for regeneration, suggesting that regenerative therapy for diabetes is feasible. Genetic lineage tracing studies indicate that β-cell regeneration is based on the replication of fully differentiated, insulin-positive β-cells. Thus, a major challenge for this field is to identify and enhance the molecular pathways that control β-cell replication and mass. We review evidence, from human genetics and mouse models, that glucose is a major signal for β-cell replication. The mitogenic effect of blood glucose is transmitted via glucose metabolism within β-cells, and through a signalling cascade that resembles the pathway for glucose-stimulated insulin secretion. We introduce the concept that the individual β-cell workload, defined as the amount of insulin that an individual β-cell must secrete to maintain euglycaemia, is the primary determinant of replication, survival and mass. We also propose that a cell-autonomous pathway, similar to that regulating replication, appears to be responsible for at least some of the toxic effects of glucose on β-cells. Understanding and uncoupling the mitogenic and toxic effects of glucose metabolism on β-cells may allow for the development of effective regenerative therapies for diabetes.