The objective of this project is to evaluate commercially available techniques for trace elemental analysis of steels using direct solid sampling. Such methods have the potential to increase sample throughput and thereby reduce analysis costs. Glow Discharge Mass Spectrometry (GD-MS) and Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICPMS) were evaluated. Five certified reference materials (CRM) were selected for the experimental work; a high purity iron, a carbon steel and three high alloy steels. For both types of instruments, the sample set was first used for calibration, then run as unknowns several times in order to get sufficient statistics for the evaluation. The GD-MS measurements were carried out at the Thermo Fisher Scientific applications laboratory in Bremen. The LA-ICPMS work was carried out at KIMAB. Prior to the measurements reported here, the effects of varying instrumental parameters of LA-ICPMS were investigated in order to develop a suitable method for steels. The results show that both techniques are powerful methods for simultaneous determination of a large number of trace elements in steels. A limitation is that none of the techniques is capable of analysing C, N and O, due to gas leaks and contamination from materials in the sources. The overall performance in terms of precision and detection limits is better for GD-MS, but for several applications LA-ICPMS provides sufficient performance. Both techniques have instrumentally limited detection limits at the ppb level or below, but the results on the test samples mostly give considerably higher standard deviations, also at very low concentration levels. Therefore, the analytical accuracy for traces is more limited than one would expect based on the instrumental capabilities; typically in the low ppm range or slightly below. The most likely reason for this is the combination of material inhomogeneity and small sampling volumes. This affects LA-ICPMS the most, due to considerably lower sampling volume than GD-MS.