All protein complexes interact with ligands during their life cycle. Identifying endogenous ligands is necessary to understand the molecular details of the complex cellular processes, but current methodologies almost universally rely on ligand extraction and separation, and thus sever the link between binding partners. This makes multiple interactions difficult to capture and unknown orphan ligands extremely challenging to identify. In this work, we combined native and omics-based mass spectrometry, to showcase a new method to identify unknown ligands directly bound to human proteins and therapeutic targets. We used multistage fragmentation to progressively dissect protein-ligand assemblies, which culminates in ligand identification via database searching. We showed that for mixtures of proteoforms and/or ligands, it is not possible to identify endogenous ligands bound to proteins either by mass measurement alone or by application of other high-resolution structural techniques such as cryoEM. To realise our workflow, we used an Orbitrap Eclipse mass spectrometer which enables transmission of high m/z ions, MSn isolation, fragmentation and simultaneous manipulation of high and low m/z species. Precursor ions up to m/z 8,000 can be routinely coupled with HCD/CID and UVPD to perform multiple stages of MS. We assessed the performance of our Multistage-nMS workflow using a variety of model proteins systems, bound to an increasingly complex cohort of endogenous ligands. We showed that performing MSn experiments in both negative and positive polarity was necessary to capture the full complement of bound species, including different families of lipids. Finally, we apply and extend our Multistage-nMS approach to identify previously unknown ligands bound to protein assemblies. The Orbitrap Eclipse capabilities offer tremendous potential to reveal how proteins assemblies are regulated in vivo, discover previously unknown endogenous ligands or signalling pathways, and to develop new avenues for novel therapeutics.