Background Although increasing the oocyte/embryo yield via an accumulation from consecutive stimulation cycles may be an effective approach towards improving IVF outcomes in POR patients, there are no robust data indicating the efficiency of the ‘accumulation scenario’ in women with a poor response undergoing IVF after multiple treatment cycles in a certain period of time. Methods This is a retrospective study of 276 infertile women in poor ovarian responders according to the Bologna criteria undergoing IVF between October 2016 to March 2020. Infertile women fulfilling the Bologna criteria in which at least two of the following three features (age ≥ 40 years, a previous POR, an abnormal ovarian reserve test) undergoing IVF were included. They used either embryo pooling (study group, n = 121) or conventional non-pooling strategy (control group, n = 155) at the discretion of the attending physicians. The primary outcome measure was the cumulative live birth rate (CLBR) of which the ongoing status had to be achieved within 24 months since ovarian stimulation. LBR was calculated by including the first live birth generated during the 24 months period whether via fresh or FET cycles. Results Implantation rate was significantly lower in the study group compared to that in the control group(17.4%vs27.8%, P = 0.002). The CLBR after multiple IVF stimulation cycle and subsequent fresh embryo transfer and /or FET cycles within 24 months follow-up were significantly lower in the study group compared to that in the control group (31.4% (38/121) and 43.2% (67/151), RR = 0.727; 95%CI: 0.528-1.000). The average time from ovarian stimulation to live birth was significantly longer in the study group compared to the control group (20.5(15.4–25.8) months vs.16(14.5–22.6) months respectively, P < 0.001). In Kaplan-Meier analysis, the cumulative incidence of ongoing pregnancy leading to live birth was significantly lower in the study group compared with control group (Log rank test, chi-square = 21.958, P < 0.001). Cox proportional hazard modelling revealed that the embryo pooling (HR = 1.896; 95% CI: (1.114–3.229), P = 0.018), age of women, number of stimulation cycles per woman (HR = 0.900; 95% CI:(0.849–0.955), P < 0.001) were strongly associated with the CLBR after adjusting other confounding factors. Conclusions The embryo pooling strategy can achieve significantly lower cumulative live birth rate when compared with the conventional non-pooling strategy. Moreover, the time to live birth was significantly longer in the embryo pooling group than the non-pooling group.