In this work, we investigated experimentally the various factors influencing the extraction of indentation stress-strain curves from spherical nanoindentation on metal samples using two different tip radii. In particular, we focused on the effects of (i) the surface preparation techniques used, (ii) the presence of a surface oxide layer, and (iii) the occurrence of pop-ins at the elastic-plastic transition on our newly developed data analysis methods for extracting reliable indentation stress-strain curves. Rough mechanical polishing was shown to introduce a large scatter in the measured indentation yield strengths, whereas electropolishing or vibropolishing produced consistent results reflective of the pristine sample. The data analysis techniques used were able to discard the portions of the raw data affected by a thin oxide layer, present on most metal surfaces, and yield reasonable indentation stress-strain curves. Experiments with different indenter tip radii on annealed and cold-worked samples indicated that pop-ins are caused by delayed nucleation of dislocations in the sample under the indenter.