Due to methodological advances in the archaeological sciences, an increasing number of archaeological specimens undergo destructive sampling. However, the preservation of cultural heritage is a primary concern. This leads to a dilemma between accessing sample material and obtaining sufficient information for a meaningful analytical outcome. Ideally, sampling a specimen would preserve the object for further macro, micro, and molecular analyses. For palaeoproteomics, a number of minimally invasive sampling approaches have been proposed, representing different benefits and limitations. There have been studies comparing a selection of these protocols, however, these have focused on specimens from a homogenous preservation environment using Zooarchaeology by Mass Spectrometry (ZooMS). Here we expand on earlier work by extending the comparison to specimens from two highly different preservation environments through both ZooMS and liquid-chromatography tandem mass spectrometry (LC-MS/MS). We compare five sampling approaches and seven extraction protocols in total, on 10 Bos sp. humeri from the Early Neolithic site of La Draga, Spain, utilising MALDI-ToF MS and LC-MS/MS to generate proteomic output, while assessing protocol invasiveness using microscopy and 3D imaging. Five humeri originate from Sector A, which is mostly related to dry, terrestrial preservation conditions, while the other five humeri stem from Sector B, which is characterised by its phreatic/aquatic preservation conditions. We show that there is a significant difference in protein recovery and taxonomic specificity between the sampling techniques applied, as well as between burial conditions. Additionally, various surface modifications were observed depending on the specific sampling technique applied. It is therefore essential to assess protein preservation for each sedimentological context within an archaeological site before performing extensive sampling, as protein preservation can be highly inter- and intra-site-specific.