The complete blood count (CBC) is used to evaluate health status in the contexts of various clinical situations such as anemia, infection, inflammation, trauma, malignancies, etc. To ensure safe clinical application of the CBC, reliable biological variation (BV) data are required. The study aim was to define the BVs of CBC parameters employing a strict protocol. Blood samples, drawn from 30 healthy subjects (17 females, 13 males) once weekly for 10 weeks, were analyzed using a Sysmex XN 3000 instrument. The data were assessed for normality, trends, outliers and variance homogeneity prior to coefficient of variation (CV)-analysis of variance (ANOVA). Sex-stratified within-subject (CV I ) and between-subjects (CV G ) BV estimates were determined for 21 CBC parameters. For leukocyte parameters, with the exception of lymphocytes and basophils, significant differences were found between female/male CV I estimates. The mean values of all erythrocyte-, reticulocyte- and platelet parameters differed significantly between the sexes, except for mean corpuscular hemoglobin concentration, mean corpuscular volume and platelet numbers. Most CV I and CV G estimates appear to be lower than those previously published. Our study, based on a rigorous protocol, provides updated and more stringent BV estimates for CBC parameters. Sex stratification of data is necessary when exploring the significance of changes in consecutive results and when setting analytical performance specifications. © 2018 Walter de Gruyter GmbH, Berlin/Boston.