As parts built through additive manufacturing (AM) increase in complexity, a strong understanding of the effect the as-built surface has on performance will be required. Currently, however, limited correlations between the as-build AM surface finish and part performance exist. This scarcity may stem not only from the complex AM build process, which leads to difficulty in isolating surface finish as a variable in functional correlation studies, but also from an inadequate consideration of surface characterization methods. In this work, a set of artifacts was built using a commercially available laser powder bed fusion system in nickel superalloy 625. From these artifacts, 648 unique surfaces were measured using a focus variation microscope. Surface height data was then characterized using the areal parameters and advanced filtering techniques available in the ISO and ASME standards and machine learning techniques were applied to identify correlations to surface position and orientation within the build chamber.