It is fundamentally important for diabetic patients to maintain an appropriate balance of diet and exercise, although a clear way to achieve that goal has not yet been established. To help achieve it, we developed a novel smartphone-based self-management support system for type 2 diabetic patients. This new system automatically stratifies each patient's biometric data according to medical risk-evaluation criteria. The data—blood glucose, blood pressure, food intake and exercise—are measured at home by the patients with devices whose sensors automatically send the data to the DialBetics server. The system’s stratification engine feeds back the raw data and the readings’ risk levels to the patients. The readings are also sent to the administrator—but only if the risk levels are so high that patients must be urged to consult their health professionals as soon as possible. Even though the efficacy of telemedicine is generally granted, the fear has been that it may increase the burden on healthcare workers, making it difficult to maintain and promote. This new system was developed specifically to overcome this kind of apprehension, actually reducing the burden on healthcare workers. We conducted a clinical study with 56 diabetes patients, its protocol approved by the ethics committee. A full paper reported the safety and efficacy of DialBetics in the Journal of Diabetes Science and Technology. We also promote the system in collaboration with a research laboratory in the Faculty of Engineering. We plan to conduct a new, more comprehensive clinical study as an advanced medical technology program at the University of Tokyo to obtain healthcare coverage of the system.