Advances in the ultra-high-resolution mass spectroscopy lead to a deep insight into the molecular characterization of natural organic matter (NOM). Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS) has been used as one of the most powerful tools to decipher NOM molecules. In FTICR-MS analysis, the matrix effects caused by the co-occurring inorganic substances in water samples greatly affect the ionization of NOM molecules. The inherent complexity of NOM may hinder its component classification and formula assignment. In this study, basic principles and recent advances for sample separation and purification approaches, ionization methods, and the evolutions in formula assignment and data exploitation of the FTICR-MS analysis were reviewed. The complementary characterization methods for FTICR-MS were also reviewed. By coupling with other developed/developing characterization methods, the statistical confidence for inferring the NOM compositions by FTICR-MS was greatly improved. Despite that the refined separation procedures and advanced data processing methods for NOM molecules have been exploited, the big challenge for interpreting NOM molecules is to give the basic structures of them. Online share of the FTICR-MS data, further optimizing the FTICR-MS technique, and coupling this technique with more characterization methods would be beneficial to improving the understanding of the composition and property of NOM.