The ISO standard regulating gear-rolling measurement does not specify in detail the calibration and verification procedures for this type of equipment. This may be one of the reasons for the lack of reproducibility in these rolling tests. The uncertainty budget method, which is the most appropriate way to know the accuracy of this dynamic measurement, shows that the measuring sensors’ accuracy is only a part of the total measurement process uncertainty. In this work, a new calibration and verification procedure for a worm gear rolling tester is presented, based on machine tool, coordinate measuring machine and gear measuring instruments’ calibration techniques. After compensating numerically for the measuring instruments, it has been evaluated how the error components of each movement affect the meshing point, a fundamental factor to ensure a good gear transmission. The study shows that there are unintentional position variations, not detected by the measuring sensors, that have to be identified and quantified in the calibration for their later inclusion in the uncertainty budget. In this way, the measurement uncertainty could be reduced, and thus improve the reproducibility of these testers, as a preliminary stage to the development of optimized rolling measurement equipment to solve current limitations.