Geometric monitoring of slender structures in power transmission infrastructure, particularly steel lattice towers of high-voltage lines, is an important engineering surveying task due to their susceptibility to environmental loads and their role in operational safety. This study compares the determination of the geometric axes of lattice transmission towers using classical tachymetric measurements and terrestrial laser scanning (TLS) performed with a compact Leica BLK360 scanner. The objective was to assess the consistency of these two methods in determining planimetric axis deviation, and to evaluate how the quality of point cloud registration impacts the results. The analysis was carried out at fourteen height levels, with deviations referred to the lowest measurement level. The TLS data were evaluated using registration quality parameters (overlap, strength, cloud-to-cloud, and global bundle error) obtained in Cyclone REGISTER 360 PLUS. Registration was performed in cloud-to-cloud mode without the use of control points. The results revealed significant discrepancies between the two methods, with local deviations exceeding 100 mm. The maximum planimetric axis deviation derived from TLS reached 140 mm, whereas the corresponding value obtained from tachymetry did not exceed 44 mm at the same inclination. Increasing differences were observed in the upper parts of the structure. These findings suggest that TLS measurements taken without stable geometric control cannot be relied upon to accurately assess axis deviation. In practice, combining the geometric stability of tachymetric measurements with the spatial completeness of TLS data offers the most effective solution for precise axis evaluation and comprehensive geometric documentation.