An expression is derived that relates the signal-to-noise (S/N) ratio in continuous-mode electrophoresis to system parameters. In this process, the time-based cross-correlation function (TCCF) is computed from signals caused by individual analyte passage through two separate detectors. As a characteristic of the noise, the cross-variance of the TCCF is proposed, so that the S/N ratio is defined as the TCCF divided by the square root of the cross-variance. From the derived expressions, the important parameters for method optimization are identified. The dependence of the S/N ratio on system parameters varies with analyte size due to a contribution from the variance in the flight time, which is affected by particle diffusion. For all molecules, the S/N increases with the processing time and the rate at which particles pass through the detectors. The latter increases with particle concentration and migration velocity. Lower sample concentrations require a longer processing time to achieve a specified S/N ratio. For smaller analytes, the residence time in the detector becomes a factor and, therefore, so does detector geometry.