In this publication, we report on the development of a quantitative enzymatic method for the detection of four botulinum neurotoxin (BoNT) serotypes responsible for human botulism by MALDI-TOF mass spectrometry. Factors that might affect the linearity and dynamic range for detection of BoNT cleavage products were initially examined, including the amount of peptide substrate and internal standard, the timing of cleavage reaction, and the components in the reaction solution. It was found that a long incubation time produced sensitive results, but was not capable of determining higher toxin concentrations, whereas a short incubation time was less sensitive so that lower toxin concentrations were not detected. In order to overcome these limitations, a two-stage analysis strategy was applied. The first stage analysis involved a short incubation period (e.g., 30 min). If no toxin was detected at this stage, the cleavage reaction was allowed to continue and the samples were analyzed at a second time point (4 h), so that toxin levels lower than 1 mouse LD50 or 55 attomoles per milliliter (55 amol/mL) could be quantified. By combining the results from two-stage quantification, 4 or 5 orders of magnitude in dynamic range were achieved for the detection of the serotypes of BoNT/A, BoNT/B, BoNT/E, or BoNT/F. The effect of multiplexing the assay by mixing substrates for different BoNT serotypes into a single reaction was also investigated in order to reduce the numbers of the cleavage reactions and save valuable clinical samples.