Defining the molecular genetics of diabetes gives new insight into the underlying etiology and so should help improve treatment. The genetic etiology is now known for most patients with β-cell monogenic diabetes, allowing genetic classification. We review how this genetic knowledge alters treatment. Patients with a glucose-sensing β-cell defect due to glucokinase mutations have regulated, mild, fasting hyperglycemia. Oral hypoglycemic agents or low-dose insulin rarely improve glycemic control. Patients with hepatic nuclear factor-1α (HNF1α) mutations have progressive β-cell deterioration and require treatment. HNF1α patients are 4 times more sensitive to sulfonylureas than matched type 2 diabetic patients. This is partly due to greater insulin secretion, reflecting the fact that the defect in HNF1α deficiency precedes the KATP channel where sulfonylureas act. HNF1β is expressed in pancreatic stem cells before differentiation into endocrine or exocrine cells, so patients with HNF1β mutations have reduced pancreatic development, resulting in early-onset diabetes and exocrine dysfunction. These patients usually rapidly require insulin and are not sensitive to sulfonylureas. Thirty-five to 50% of patients diagnosed with diabetes before 6 months have a mutation in Kir6.2. The mutated KATP channel in these patients does not close in response to increased ATP concentrations, but can be closed when sulfonylureas bind to the sulfonylurea receptor 1 subunit of the channel by an ATP-independent route. These patients are usually insulin dependent, but have excellent glycemic control on high-dose sulfonylureas tablets. In conclusion, the defining of molecular genetic etiology in monogenic diabetes has identified several specific β-cell defects, and these are critical in determining the response to treatment.