Introduction: Echocardiography is widely used to optimize CRT programming, but it is time‐consuming. This study aimed to correlate the optimal interventricular pacing (V‐V) interval obtained by echo with the optimal V‐V interval obtained by a simpler method based on the surface ECG. Methods and Results: Three V‐V intervals were tested: LV preactivation at –30 ms, simultaneous biventricular pacing (0 ms), and RV preactivation at +30 ms. The one that achieved the best LV synchrony was chosen as the optimal V‐V. This result was then compared with two different ECG measurements. The first ECG method considered the best V‐V to be that which achieved the narrowest QRS. The second V‐V method consisted in measuring the interval from the pacing spike to the beginning of the fast deflexion of the QRS complex in leads V1, V2, first pacing from the LV (T1), and after from the RV (T2). The T2‐T1 interval was considered as a surrogate measurement of interventricular delay and defined as the best V‐V. A cohort of 31 consecutive patients treated with CRT was studied. Optimal V‐V interval obtained by echo was –30 ms in 25 patients (80%), +30 ms in three patients (10%), and 0 ms in the remaining three patients (10%). Echo results had 32% coincidence with the first ECG method (r = 0.2, P = NS) and 83% coincidence with the second ECG method (r = 0.81 P< 0.001). Conclusions: The time difference in the fast ventricular depolarization observed between RV and LV stimulation in the surface ECG shows a good correlation with the V‐V optimization chosen according to echo.