Abstract Study question Could embryo accumulation employing mild stimulation cycles prove beneficial for managing patients presenting with poor ovarian response (POR)? Summary answer Embryo accumulation may be an efficient POR management strategy, enabling a higher number and quality cohort of embryos, ultimately improving success results. What is known already It is widely accepted that POR constitutes a challenging condition. The limited oocyte yield associated with POR detrimentally impacts in vitro fertilization (IVF) success rates. Moreover, the documented heterogeneity among POR patients compromises our efforts to successfully address POR, despite the advances noted regarding stimulation protocols employed today. Considering the aforementioned, embryo accumulation following consecutive stimulation cycles has emerged as an alternative management strategy towards increasing the number of available embryos prior to embryo transfer (ET), mimicking normoresponding conditions. However, only few studies have been so far conducted and the need for further data is underlined. Study design, size, duration A single-center retrospective study was conducted in the Centre of Human Reproduction, Genesis-Athens Clinic from January 2015-December 2019. Only patients presenting with POR according to Bologna criteria were included. In total, 610 POR patients were considered eligible and were divided in three groups namely, mild stimulation-fresh ET (150 IUs of gonadotropins) (MILDF), mild stimulation employing embryo accumulation (MILDA), and natural cycle employing embryo accumulation (NATA). Respective comparisons on embryology and pregnancy data are provided. Participants/materials, setting, methods Resulting embryos from the MILDF, MILDA, and NATA groups were cultured up to the cleavage stage and categorized into three groups according to quality, namely top (grade 1), good (grade 2–3) and poor (grade 4–5) (Veeck, 1999). Top and good quality embryos were considered eligible for ET/vitrification. The banking scenario entailed accumulation of at least three embryos, including at least one top quality embryo. Embryo transfers included up to two cleavage stage embryos. Main results and the role of chance Comparing MILDF and MILDA groups, a higher number of available oocytes and embryos was observed in MILDA (2.36±1.15 vs 6.58±1.11; 1.72±1.02 vs 3.51±0.61, P-value<0.001). However, a mean number of 3.90±1.56 oocyte retrievals were required to conclude MILDA compared to MILDF which was concluded following a single oocyte retrieval (P-value<0.001). Cancellation-rate was significantly lower in the MILDA compared to MILDF group (0% vs 18.93%, P-value <0.001). A higher proportion of top quality embryos were transferred in the MILDA group (66.58% vs 43.67%, P-value<0.001). The MILDA group presented with higher positive-HCG (27.89% vs 23.30%, P-value=0.302), clinical-pregnancy (22.11% vs 17.96%, P-value=0.316) and live-birth rates (16.84% vs 14.08%, P-value=0.487). However, these differences were not significant. Comparing MILDA and NATA groups, the MILDA presented with a lower number of required oocyte retrievals and a higher number of oocytes per oocyte retrieval compared with NATA (3.90±1.56 vs 7.15±1.80; 1.95±0.74 vs 0.89±0.20, P-value<0.001). Moreover, the MILDA presented with a higher mean number of resulting embryos (5.20±0.78 vs 4.82±0.88, P-value<0.001). No difference was observed regarding the proportion of the resulting top quality embryos. The MILDA group presented with slightly higher clinical-pregnancy (22.11% vs 20.09%, P-value=0.628) and live-birth (16.84% vs 14.02%, P-value=0.490) rates, however these differences were not significant. Limitations, reasons for caution The retrospective nature of the study constitutes a major limitation. Considering that numerous confounders are inevitable when retrospective data is analyzed, authors employed strict eligibility criteria in an effort to reduce bias. Statistical analysis revealed a well-controlled population, considering that general patients’ characteristics did not differ between the three groups. Wider implications of the findings: Embryo accumulation may constitute an efficient management strategy for POR, as more embryos of better quality are available for ET compared to fresh-IVF-ET. Mild stimulation should be preferred for embryo accumulation instead of natural cycles, as less oocyte retrievals are required. Future studies should be conducted to verify these conclusions. Trial registration number Not applicable