Metabolic profiling of tissues needs special attention, because the compartmentalization of cellular constituents will be abolished by sample homogenization. This loss of partitioning leads to protein and metabolite instability in extracts, and therefore metabolite extraction protocols need to ensure very rapid inactivation of mac- romolecules as well as solubilization of metabolites. There are many published methods for tissue metabolome anal- ysis, but no universally accepted standard, and a lack of measurable quality benchmarks. We developed a protocol for efficient tissue disruption and metabolite extraction of the earthworm Lumbricus rubellus guided by prior biological knowledge as well as metrics based on the data. In particular, we identified an unusual degree of instability of L. rubellus tissue extracts, and evaluated different approaches such as heating and filtration to counteract this. Finally, we evaluated four different solvent systems for comprehensive metabolite extraction using three analytical platforms (1H NMR spectroscopy, GC–MS, and direct- infusion FT-ICR-MS), and also compared bead-beating and cryogenic milling for tissue disruption. Initially we ranked methods by common analytical criteria (e.g. num- bers and total intensity of detected peaks) in order to compare protocols. These approaches to assess protocol suitability proved to be inadequate to judge earthworm tissue extraction methods because of sample instability. Existing tissue extraction protocols should not be assumed to be automatically applicable to novel species.