High performance liquid chromatography coupled with quadrupole-time-of-flight mass spectrometry (HPLC-QTOF-MS) is widely used to qualitatively characterize the chemical profiles of herbal medicines, in which the generated adducts and fragments are crucial for confirming molecular ion (deprotonated/protonated ion) and deducing structure of detected components. However, how chromatographic and mass spectrometric (LC–MS) conditions/parameters affect the quantity and intensity of adducts and fragments of detected components is scarcely concerned. In present study, three types of triterpene saponins from the root of Ilex asprella (RIA) were selected as a case study to systematically investigate the effects of LC/MS conditions/parameters on their ionization and fragmentation, so as to obtain higher intensity (higher detection sensitivity) and quantity (rich information) of adducts and fragments for the characterization of components in RIA. It was found that for LC conditions, methanol as organic phase was more benefit for generating more adducts with higher intensity; formic acid as a modifier suppressed the formation of [M-2H], thus promoted the generation of other types of adducts at lower concentration but inhibited the generation when the concentration exceeded 0.1%. MS parameters affect scarcely the quantity but mainly intensity of adducts, cone voltage, source temperature and desolvation gas flow have relatively higher impacts when compared with other parameters. Collision energy affected both quantity and intensity of fragments. MS parameters at the medium value largely increased the quantity and intensity of adducts and fragments. Three-types of triterpene saponins presented structurally specific ionization and fragmentation due to their amounts of acidic substitutes. A total of 55 components were detected and definitely or tentatively identified in RIA under the optimized LC–MS conditions, among which 35 triterpene saponins were firstly discovered. This is the first report that proposes and validates a systematic approach for assessing the effects of LC/MS conditions/parameters on the ionization and fragmentation of analytes, which could be helpful for the optimization of LC–MS conditions for effective chemical profiling analysis of herbal medicines.