Introduction: Cardiac resynchronization therapy (CRT) is one of the most important advances in heart failure management in the last twenty years. Approximately one-third of patients appear not to respond to therapy. Although there are a number of possible mechanisms for non-response, an important factor is suboptimal atrioventricular (AV) and interventricular (VV) timing intervals. There remains controversy over whether routinely optimizing intervals is necessary and there is no agreed gold standard methodology. Optimization has classically been performed using echocardiography which has limits related to resource use, time-cost and variable reproducibility. Newer optimization methods using device-based sensors and algorithms show promise in reducing heart-failure hospitalization compared with echocardiography.Areas covered: This review outlines the rationale for optimization, the principles of AV and VV optimization, the standard echocardiographic approach and newer device-based algorithms and the evidence base for their use.Expert commentary: The incremental gains of optimization are likely to be real, but small, compared to the overall improvement gained from cardiac resynchronization itself. At this time routine optimization may not be mandatory but should be performed where there is no response to CRT. Device-based optimization algorithms appear to be practical and in some cases, deliver superior clinical outcomes compared to echocardiography.