Background With the development of cord blood banking, solutions have to be found to solve the storage space problem, by reducing the volume of cord blood units (CBU). Methods We compared total nucleated cell (TNC) and CD34(+) cell counts before and after processing with three different CBU volume reduction methods used consecutively in our bank: a manual method based on hydroxyethyl starch sedimentation (HES) (n=447), a top-and-bottom (TB) semi-automated method (n=181) using Optipress II, and the Sepax automated method (n=213). Statistical analysis was done using t-tests, linear regression and Spearman correlation coefficients. Adjusted variables included TNC, CD34(+) cell counts, CD34(+) cell percentage and CB volume before processing. Results TNC recovery was higher with Sepax (80.3+/-7.7%) than with HES (76.8+/-9.1%) and TB (60.7+/-13.5%) (P<0.0001, both). It was higher with HES than with TB (P<0.0001). CD34(+) cell recovery was higher with Sepax (86+/-11.6%) than with HES (81.5+/-12.5%) and TB (82.0+/-17.7%) (P<0.008 and <0.0001, respectively) and results with HES and TB were not significantly different (P=0.7). Interestingly, with Sepax, TNC and CD34(+) cell recoveries were not correlated with pre-processing values (P=0.8 and 0.4, respectively). Discussion In conclusion, the Sepax volume reduction method allows higher TNC and CD34(+) cell recoveries.