Uncertainty evolved through in situ testing, such as measurement error, is a major concern in the characterization of soft clays that suffer from the low strength and high sensitivity of soil properties. In this research, 20 piezocone penetration tests (CPTUs) in a laboratory calibration chamber and 25 CPTUs performed in marine soft clay in Lianyungang, Jiangsu province, China, are considered for characterizing various in situ measurement errors. Equipment error, calibration error, penetration pause effect, and random measurement error are identified as the four main sources of measurement errors based on comprehensive literature review and field observations. Laboratory studies revealed that a coefficient of variability (COV) of 4 % was recorded for uncertainties associated with calibration of the device. Analyses performed on the in situ measurements showed that the COVs of penetration pause effects on cone tip resistance values varied from 0 % to 8 %. The methods of moments and maximum likelihood approach were used to estimate the random measurement error of qt profiles based on the analysis of correlation structure. Both methods showed that COVs of random measurement error of qt ranged from 0 % to 12 %. Using the second-moment statistics, the COVs of the total measurement errors comprising of calibration and equipment error, penetration pause effects and random measurement errors were estimated to vary from 5 % to 15 %. Consequently, a representative CPTU sounding example was used to demonstrate the influence of measurement error on the tip resistance measurements. Overall, this research presents a rational method to improve the accuracy of CPTU measurements in marine soft clay by separating the uncertainties associated with measurement errors.