Neonatal Septicemia is a serious clinical syndrome and the definitive diagnosis is based on positive blood cultures which are obtained either by conventional or automated method. Early availability of proper isolation and identification of causative bacteria facilitates the timely initiation of appropriate antibiotic therapy. Thereby the present study was conducted to identify the bacterial causes of neonatal septicemia in the fastest possible time by comparing conventional and automated blood culture methods.This cross sectional study was done during the period from January 2018 to December 2018 and included clinically suspected cases of neonatal septicemia admitted to Neonatal Intensive Care Units of Chattogram Medical College Hospital (CMCH) and Chattogram Maa-Shishu O General Hospital (CMSOGH). Out of 178 samples, automated method detected 29 (16.3%) and conventional method detected 26 (14.6%) blood culture positive samples. The yield of bacteria by automated method was 100% and by conventional method was 89.7%. Number of bacteria isolated only by automated method were 3 (10.2%). Mean time for isolation of bacteria by automated method was 26.38 hours and by conventional method was 46.34 hours. Automated method detected 47.05% of isolated bacteria in first 24 hours but none of them were detected by conventional method within first 24 hours. Among the isolated bacteria, Klebsiella spp was most common (62.0%). Most of the isolates were resistant to Ampicillin, Cefotaxime and Ceftazidime. Analyzing the findings of the study, there was no significant difference in the rate of isolation in each time interval (p=0.157) of two methods but there was significant difference in the mean time of isolation of bacteria between two methods (p=0.000004). Bangladesh J Med Microbiol 2021; 15 (2): 12-18