We conducted a prospective observational study to evaluate the medium-term impact of the flash glucose monitoring system (FGM) in a type 1 diabetic population. We included 248 patients, switched from conventional blood glucose monitoring (BGM) to FGM. We evaluated glycaemic control at 2-4 (T1) and 5-11 (T2) months after initiation and at the last available visit (T3, 18 ± 4 months). We asked patients to fill in, at T0 and T2, two questionnaires based on the Diabetes Treatment Satisfaction Questionnaire; and on the Hypoglycaemia Fear Survey. Glycaemic control improved, from 8.1 ± 1.3% at T0 to 7.8 ± 1.2% at T1 (p < 0.001) and remained unchanged after. Average number of controls increased from 3.2 ± 1.2 BGM to 7.7 ± 3.9 at T1 (p < 0.001). We observed a modest decrease in daily insulin doses. We evidenced an increase in mild hypoglycaemic events, especially in well-controlled subjects, but no increase of severe events. Satisfaction score improved from 30.5 ± 7.7 points to 38.3 ± 5.1 points (p = 0.018), was correlated with the reduction in and was higher in less controlled patients at inclusion. "Behaviour" score regarding hypoglycaemias decrease from 5.7 ± 4.1 to 4.4 ± 3.6 points (p < 0.001). In conclusion, this 18-months study trial indicates that using the FGM technology in patients with T1D may improve glycaemic control, in real-life conditions, especially in less controlled patients. FGM was associated with an increase of patients' satisfaction regarding treatment. Hypoglycaemic events, however, were not reduced in frequency. Therefore, the need for an educational team and a structure program in the management of this new technology remains mandatory.