According to GB 4789.35-2016 National Food Safety Standard， Food microbiology examination： Lactic acid bacteria， the dilution and add sample can be conducted with two different kinds of tools， the pipette or the micropipettor. However， using these two different tools whether significantly affects the uncertainty of counting result， when counting Bifidobacterium of health food samples. The uncertainty of Bifidobacterium count of health food samples is measured in accordance with Evaluation and Expression of uncertainty in Measurement， publish by People's Republic of China State Administration of Quality Supervision， Inspection and Quarantine， JJF 1059.1-2012. The uncertainty is evaluated from the perspectives of the sample preparation， sample dilution， sample addition and the repetitive test， and using the pipette or the micropipettor in testing. For the sample diluted and added by the pipette， the uncertainty level U is 0.28 under 95% confidence level. The counted number for the Bifidobacterium is （2.2-8.1）×109 CFU/g which is equal to 9.35-9.91 after the natural logarithm. The confidence interval for the sample diluted by pipette is （2.2-8.1）×109 CFU/g. As for the sample diluted and added by the micropipettor， the uncertainty level U is 0.36 under 95% confidence level. The counted number for the Bifidobacterium is （2.7-14）×109 CFU/g which is equal to 9.43-10.15 after the natural logarithm. The confidence interval for the sample diluted by micropipettor is （2.7-14）×109 CFU/g. The uncertainty of count of Bifidobacterium in health food sample is about 90% caused by sample dilution， because the count of Bifidobacterium is huge， and dilution factor is big. The uncertainty by using pipette is a little bit smaller than using micropipettor， but a little bit bigger in the uncertainty A. The confidence interval using pipette is a little bit smaller than using micropipettor. Specially， the result of Bifidobacterium counting is significant differences in the same sample， using pipette and micropipettor.