BACKGROUND: Clinical transplant outcome with umbilical cord blood (UCB) as source of hematopoietic progenitor cells (HPCs) is, among other factors, determined by the total number of viable nucleated cells and/or CD34+ cells in the unit. Quantitative and qualitative losses by processing and cryopreservation and by thawing and washing before transfusion may occur, however. Another reason for a discrepancy between the number of cells in the unit released by the cord blood bank and found in the transplant center may be technical differences in cell counting methods between the two sites. STUDY DESIGN AND METHODS: With the collaborative group for Biomedical Excellence for Safer Transfusion (BEST), an interlaboratory exercise was conducted among nine sites for thawed UCB variables: total nucleated cells, CD34+ cells, viability, and HPC cultures. Three frozen UCB samples were shipped, with instructions for thawing, counting, and HPC plating. RESULTS: Unexpectedly samples arrived at all nine receiving centers without detectable hematopoietic progenitor colony-forming cells. Nevertheless, wide interlaboratory ranges for viability were obtained. The proportion of viable cells was found higher with manual methods, but all viability assays used in the study overestimated functional progenitor cells. CONCLUSIONS: The results underscore the complexity of evaluation of frozen-thawed cord blood cells and the need for standardization of assessment.