Cardiomyocytes, the contractile cells of the heart, are known to be polyploid, meaning they have undergone whole-genome duplication(s) and contain more than two complete sets of chromosomes. However, the exact functions of cardiomyocyte polyploidy are unknown. In this study, Archan Chakraborty, Donald Fox and colleagues characterise polyploidy of Drosophila cardiomyocytes across embryonic and larval stages of development. Here, they find that levels of ploidy differ between the anterior aorta and posterior heart chambers. In addition, in cardiomyocytes from human organ donors, they find a similar difference in ploidy between left atrium and ventricle chambers. By analysing RNA-seq datasets from human atrial and ventricular cardiomyocytes, the authors identify a positive correlation between ploidy levels and expression of insulin receptor (INSR). Indeed, blocking insulin signalling by RNAi depletion of INSR in Drosophila reduces endoreplication of cardiomyocytes and thus ploidy levels. This results in reduced chamber sizes and cardiac performance, including attenuated stroke volume and cardiac output, phenotypes resembling those of human cardiomyopathies. Thus, these findings demonstrate that chamber-specific cardiomyocyte polyploidy is important for precise heart development and function. Furthermore, developmental polyploidisation has previously been shown to block cardiac regeneration, and so these findings offer potential insights into how polyploidisation could be blocked to promote regeneration after heart injury.