Understanding the status and global trends of soil invertebrate diversity requires accurate and comparable data across geographical regions. However, soil animal extraction approaches still vary among laboratories, and no commonly accepted, openly available and well-documented protocols exist across taxa. Here, we present harmonised methodologies, assembled by an international group of experts, for extracting soil- and litter-inhabiting nematodes, enchytraeids, microarthropods, and larger invertebrates. The protocols are illustrated with images and videos, and include advice for overcoming the most frequent issues (‘expert tips’) for increasing extraction efficiency and reproducibility. In addition, we provide results from two pilot experiments on nematode and large invertebrate extractions. First, we show that using two layers of milk filters instead of one in wet extraction of nematodes yields very similar extraction efficiency and has little effect on the sample cleanliness. Second, we demonstrate that on average approximately 31.3% of large soil animals (3 mm and more in body length) are overlooked during hand sorting but were then captured in heat extraction of the sorted samples across three geographically distinct laboratories. The extent of bias varied by collection team and taxon, with particularly high collection bias for Pseudoscorpiones (64.4% overlooked), Diptera (62.9%) and Coleoptera larvae (57.1%). Observed differences among the laboratories call for standardized ring tests of extraction efficiency across regions and research groups. Overall, we provide openly available protocols for assessing soil animal taxa in ecological studies worldwide, facilitating comparisons for a better understanding of the dynamics of global soil biodiversity.