The environmental mobility of Cu and therefore its potential toxicity are closely linked to its attachment to natural organic matter (NOM). Geochemical models assume full lability of metals bound to NOM, especially under strong oxidizing conditions, which often leads to an overestimation of the lability of soil metals. Stable isotope dilution (SID) has been successfully applied to estimate the labile (isotopically exchangeable) pool of soil metals. However, its application to study the lability of NOM-Cu required development of a robust separation and detection approach so that free Cu ions can be discriminated from (the also soluble) NOM-Cu. We developed a SID protocol (with enriched 65Cu) to quantify the labile pool of NOM-Cu using size exclusion chromatography coupled to a UV detector (for the identification of different NOM molecular weights) and ICP-MS (for 65Cu/63Cu ratio measurement). The Cu isotopic-exchange technique was first characterized and verified using standard NOM (SR-NOM) before applying the developed technique to an "organic-rich" podzol soil extract. The developed protocol indicated that, in contrast to the common knowledge, significant proportions of SR-NOM-Cu (25%) and soil organic-Cu (55%) were not labile, i.e., permanently locked into inaccessible organic structures. These findings need to be considered in defining Cu interactions with the reactive pool of NOM using geochemical models and risk evaluation protocols in which complexed Cu has always been implicitly assumed to be fully labile and exchangeable with free Cu ions.