Myocardial infarction resulting in irreversible loss of cardiomyocytes (CMs) remains a leading cause of heart failure. Although cell transplantation has modestly improved cardiac function, major challenges including increasing cell survival, engraftment, and functional integration with host tissue, remain. Embryonic stem cells (ESCs), which can be differentiated into cardiac progenitors (CPs) and CMs, represent a candidate cell source for cardiac cell therapy. However, it is not known what specific cell type or condition is the most appropriate for transplantation. This problem is exasperated by the lack of efficient and predictive strategies to screen the large numbers of parameters that may impact cell transplantation. We used a cardiac tissue model, engineered heart tissue (EHT), and quantitative molecular and electrophysiological analyses, to test transplantation conditions and specific cell populations for their potential to functionally integrate with the host tissue. In this study, we validated our analytical platform using contractile mouse neonatal CMs (nCMs) and noncontractile cardiac fibroblasts (cFBs), and screened for the integration potential of ESC-derived CMs and CPs (ESC-CMs and -CPs). Consistent with previous in vivo studies, cFB injection interfered with electrical signal propagation, whereas injected nCMs improved tissue function. Purified bioreactor-generated ESC-CMs exhibited a diminished capacity for electrophysiological integration; a result correlated with lower (compared with nCMs) connexin 43 expression. ESC-CPs, however, appeared able to appropriately mature and integrate into EHT, enhancing the amplitude of tissue contraction. Our results support the use of EHT as a model system to accelerate development of cardiac cell therapy strategies.