Capillary electrophoresis with capacitively coupled contactless conductivity detection (CE-C4D) has proven to be an efficient technique for the separation and detection of charged inorganic, organic, and biochemical analytes. It offers several advantages, including cost-effectiveness, nanoliter injection volume, short analysis time, good separation efficiency, suitability for miniaturization, and portability. However, the routine determination of common inorganic cations (NH4+, K+, Na+, Ca2+, Mg2+, and Li+) and inorganic anions (F-, Cl-, Br-, NO2-, NO3-, PO43-, and SO42-) in water quality monitoring typically exhibits limits of detection of about 0.3-1 μM without preconcentration. This sensitivity often proves insufficient for the applications of CE-C4D in trace analysis situations. Here, we explore methods to push the detection limits of CE-C4D through a comprehensive consideration of signal and noise sources. In particular, we (i) studied the model of C4D and its guiding roles in C4D and CE-C4D, (ii) optimized the bandwidth and noise performance of the current-to-voltage (I-V) converter, and (iii) reduced the noise level due to the strong background signal of the background electrolyte by adaptive differential detection. We characterized the system with Li+; the 3-fold signal-to-noise (S/N) detection limit for Li+ was determined at 20 nM, with a linear range spanning from 60 nM to 1.6 mM. Moreover, the optimized CE-C4D method was applied to the analysis of common mixed inorganic cations (K+, Na+, Ca2+, Mg2+, and Li+), anions (F-, Cl-, Br-, NO2-, NO3-, PO43-, and SO42-), toxic halides (BrO3-) and heavy metal ions (Pb2+, Cd2+, Cr3+, Co2+, Ni2+, Zn2+, and Cu2+) at trace concentrations of 200 nM. All electropherograms showed good S/N ratios, thus proving its applicability and accuracy. Our results have shown that the developed CE-C4D method is feasible for trace ion analysis in water quality control.