This is a retrospective study to evaluate if the miRNome profile of endometrium samples collected during the implantation window predicts Assisted Reproduction Technology (ART) outcomes. We first investigated the endometrial miRNome profile according to the receptivity status in 20 patients with repeated implantation failures (RIF) (discovery cohort). After customized embryo transfer, the miRNome profiles of receptive patients with a positive or negative β-hCG, and with early miscarriage or live birth were analysed. Some differentially expressed miRNAs were selected for validation by RT-qPCR in endometrial samples from 103 RIF patients (validation cohort). Analysis of the different miRNome profiles identified endometrial receptivity, implantation failure, and early miscarriage-associated miRNA signatures that included 11, 261, and 76 miRNAs, respectively. However, only four miRNAs associated with the endometrial receptivity status (miR-455-3p and miR-4423-3p) and implantation failure (miR-152-3p and miR-155-5p) were significantly validated in endometrial samples. The miRNome profile of endometrial tissues during the implantation window can predict the pregnancy outcome. These data are crucial for opening new perspectives to predict implantation failure and consequently, to increase ART success.