Aims We performed a randomized, double blind, crossover study of cardiac contractility modulation (CCM) signals in heart failure patients. Methods and results One hundred and sixty-four subjects with ejection fraction (EF) < 35% and NYHA Class II (24%) or III (76%) symptoms received a CCM pulse generator. Patients were randomly assigned to Group 1 (n = 80, CCM treatment 3 months, sham treatment second 3 months) or Group 2 (n = 84, sham treatment 3 months, CCM treatment second 3 months). The co-primary endpoints were changes in peak oxygen consumption (VO2,peak) and Minnesota Living with Heart Failure Questionnaire (MLWHFQ). Baseline EF (29.3 ± 6.7% vs. 29.8 ± 7.8%), VO2,peak (14.1 ± 3.0 vs. 13.6 ± 2.7 mL/kg/min), and MLWHFQ (38.9 ± 27.4 vs. 36.5 ± 27.1) were similar between the groups. VO2,peak increased similarly in both groups during the first 3 months (0.40 ± 3.0 vs. 0.37 ± 3.3 mL/kg/min, placebo effect). During the next 3 months, VO2,peak decreased in the group switched to sham (−0.86 ± 3.06 mL/kg/min) and increased in patients switched to active treatment (0.16 ± 2.50 mL/kg/min). MLWHFQ trended better with treatment (−12.06 ± 15.33 vs. −9.70 ± 16.71) during the first 3 months, increased during the second 3 months in the group switched to sham (+4.70 ± 16.57), and decreased further in patients switched to active treatment (−0.70 ± 15.13). A comparison of values at the end of active treatment periods vs. end of sham treatment periods indicates statistically significantly improved VO2,peak and MLWHFQ (P = 0.03 for each parameter). Conclusion In patients with heart failure and left ventricular dysfunction, CCM signals appear safe; exercise tolerance and quality of life (MLWHFQ) were significantly better while patients were receiving active treatment with CCM for a 3-month period.