Remote monitoring (RM) for cardiac implantable electronic devices is on the rise and has been shown to reduce the burden of in-clinic follow-up visits. We aimed to investigate the long-term clinical outcomes of RM versus no RM. This was a prospective, single-centre cohort study of consecutive patients with an implantable cardioverter-defibrillator (ICD) or cardiac resynchronisation therapy-defibrillator (CRT-D) followed up from 2018 to 2023. Patients who received non-ICD devices were excluded. In general, RM was offered to all patients, but uptake depended on patient preference. For data analysis, patients were stratified according to whether RM was used. The primary outcome was all-cause mortality; secondary outcomes were hospitalisation for heart failure and device therapy (shocks and electrical storm). Of 551 patients, 284 (51.5%) received RM and 267 (49.5%) did not. Baseline demographics were similar between the two arms. All-cause mortality was significantly lower in RM versus non-RM patients (hazard ratio [HR] 0.45, 95% confidence interval [CI] 0.33-0.60, P <0.001), as was hospitalisation for heart failure (HR 0.39, 95% CI 0.25-0.59, P <0.001); these remained significant after adjustment for baseline covariates. More patients on RM received appropriate antitachycardia pacing (ATP) (17.6% vs. 10.7%, P = 0.035) and appropriate shocks (24.1% vs. 14.7%, P = 0.017). The incidences of inappropriate ATP, inappropriate shocks and electrical storm were similar. More patients on RM underwent pulse generator change (34.1% vs. 10.1%, P <0.001). Remote monitoring was associated with significantly lower mortality in both ICDs and CRT-Ds and in primary and secondary indications, as well as fewer heart failure hospitalisations. This supports current guidelines recommending the use of RM in all patients with ICD or CRT-D.