Selecting reasonable matches from the database search results is crucial to mass spectrometry-based proteomics identification. However, the current score-based filter solution and decoy database methods are not effective enough to prevent all false positive and false negative selections. In this study, a systematic search strategy named iterative non-m/z-sharing (INMZS) analysis was proposed to address the problem. In the strategy, all search results were screened based on the share status of corresponding matched m/z, only the proteins that matched with exclusive m/z information were reserved for the confident matches. The researchers did further iterative search to improve the identification of minor components in a spot. Finally, identifications were confirmed by decoy database evaluation for the final phase of protein identification. Simulation and application tests of INMZS were implemented on a large human liver data set and standard protein cocktails. The result shows that INMZS plus decoy database evaluation is efficient in ensuring the confidence and sensitivity of 2-DE or similar non-shotgun based proteome identification.