“MS/MS spectrum to structure” analysis is the most challenging task for MS/MS-relied qualitative characterization. The conventional database- and computation-assisted strategies cannot reach confirmative identification, notably for isomers. Hence, an advanced strategy was proposed here through tackling the two determinant obstacles such as the transformation from elemental compositions to fragment ion structures and the linkage style amongst substructures. As typical conjugated structures, esters were measured for strategy illustration, and metabolite identification of a famous natural antioxidant namely rosmarinic acid (RosA) in rat was undertaken for applicability justification. Through programming online energy-resolved (ER)-MS for the first collision cell of Qtrap-MS device, full collision energy ramp (FCER)-MS2 spectrum was configured for [M−H]− ion of each ester to provide optimal collision energies (OCEs) for all concerned diagnostic fragment ions (DFIs), i.e. a-, b-, c-, y-, and z-type ions. The linear correlations between masses and OCEs were built for each ion type to facilitate DFIs recognition from chaotic MS2 spectrum. To identify 1st-generation fragment ions, full exciting energy ramp (FEER)-MS3 spectra were configured for key DFIs via programming the second ER-MS in the latter collision chamber. FEER-MS3 spectrum of 1st-generation fragment ion for ester was demonstrated to be identical with FEER-MS2 spectrum of certain hydrolysis product when sharing the same structure. After applying the advanced strategy to recognize DFIs and identify 1st-generation fragment ions, a total of forty metabolites (M1−M40), resulted from hydrolysis, methylation, sulfation, and glucuronidation, were unambiguously identified for RosA after oral administration. Together, the advanced bottom-up strategy hyphenating FCER-MS2 and FEER-MS3 spectra, is meaningful to strengthen “MS/MS spectrum to structure” analysis through recognizing and identifying fragment ions.