Determination of the 182W/184W ratio to a precision of ±5 ppm (2σ) is desirable for constraining the timing of core formation and other early planetary differentiation processes. However, WO3 − analysis by negative thermal ionization mass spectrometry normally results in a residual correlation between the instrumental-mass-fractionation-corrected 182W/184W and 183W/184W ratios that is attributed to mass-dependent variability of O isotopes over the course of an analysis and between different analyses. A second-order correction using the 183W/184W ratio relies on the assumption that this ratio is constant in nature. This may prove invalid, as has already been realized for other isotope systems. The present study utilizes simultaneous monitoring of the 18O/16O and W isotope ratios to correct oxide interferences on a per-integration basis and thus avoid the need for a double normalization of W isotopes. After normalization of W isotope ratios to a pair of W isotopes, following the exponential law, no residual W−O isotope correlation is observed. However, there is a non-ideal mass bias residual correlation between 182W/iW and 183W/iW with time. Without double normalization of W isotopes and on the basis of three or four duplicate analyses, the external reproducibility per session of 182W/184W and 183W/184W normalized to 186W/183W is 5−6 ppm (2σ, 1−3 μg loads). The combined uncertainty per session is less than 4 ppm for 183W/184W and less than 6 ppm for 182W/184W (2σm) for loads between 3000 and 50 ng.