Abstract Study question In consecutive intracytoplasmic sperm injection (ICSI) cycles, is embryonic development in EmbryoScope better than the previous one obtained in a benchtop (G–185) incubator? Summary answer Embryonic development, and oocyte and embryo utilization rates (OUR and EUR) are significantly improved in the EmbryoScope, as compared to G–185. What is known already The time-lapse imaging (TLI) system, which allows a non-invasive continuous assessment of embryo morphokinetics parameters in a closed culture system has been developed, promising improved embryo development by reducing oscillations in pH, humidity and temperature. To investigate this hypothesis, one study has already compared embryonic development in a TLI versus a benchtop incubator. However, It has never been investigated whether embryonic development can be improved within-subject, by changing from benchtop incubator in the first intracytoplasmic sperm injection (ICSI) cycle to the EmbryoScope, a TLI incubator, in the following ICSI cycle, and that was the objective of the present study. Study design, size, duration This study had a retrospective within-subject design, in which each cycle served as its own control. Data were obtained via chart review of patients undergoing ICSI in a private university–affiliated IVF center that fulfilled the following criteria: second ICSI attempt in which embryos had been cultured in a TLI incubator system (TLI group, n = 71), preceded by a first ICSI attempt in which embryos had been cultured in a conventional incubator (Control group, n = 71). Participants/materials, setting, methods Embryonic development up to the fifth day of development, OUR (transferred embryos plus frozen embryos per retrieved oocytes) and EUR (transferred embryos plus frozen embryos per fertilized oocytes) were compared between the groups using generalized linear models followed by Bonferroni post hoc. The post hoc achieved power was 82.6%, considering the sample size, the effect size obtained for blastocyst development rate and 5% significance level. Main results and the role of chance There were significant differences in fertilization rate (76.0% ± 1.3 vs. 80.0% ± 1.4, p = 0.044, OR: 1.051, CI: 1.001 – 1.103), non-fertilization rate (14.8% ± 0.6 vs. 6.3% ± 0.4, p < 0.001, OR: 0.424, CI: 0.370 – 0.486), day–2 non-cleavage rate (3.8% ± 0.2 vs. 1.1% ± 0.1, p < 0.001, OR: 0.285, CI: 0.234 – 0.347), blastocyst development rate (40.9% ± 1.1 vs. 55.6% ± 1.3, p < 0.001, OR: 1.358, CI: 1.267 – 1.456), frozen blastocyst rate (31.8% ± 0.8 vs. 37.0% ± 0.9, p < 0.001, OR: 1.163, CI: 1. 085 – 1.248), OUR (40.7% ± 1.0 vs. 50.2% ± 1.1, p < 0.001, OR: 1.232, CI: 1.155 – 1.314), and EUR (52.4% ± 1.1 vs. 66.6% ± 1.2, p < 0.001, OR: 1.269, CI: 1.202 - 1.341), all in favor of TLI group. Pregnancy rate (30.2% vs. 30.8%, p = 0.940), implantation rate (24.6% ± 40.0 vs. 26.1% ± 41.6, p = 0.830), and miscarriage rate (21.1% vs. 15.0%, p = 0.622) were similar between Control and TLI groups, respectively. Limitations, reasons for caution (i) Different culture dishes were used in each system; (ii) it is not possible to confirm how much of the embryonic improvement was due to the culture conditions; (iii) the study design is not ideal for the comparison of clinical outcomes and, also, underpowered to do so. Wider implications of the findings: Even though the clinical outcomes were similar between the groups, the results may also lead to higher cumulative pregnancy outcomes following embryo thawing and transfer. Trial registration number Not applicable