Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry is a powerful tool for rapidly generating protein expression data (peptide and protein profiles) from a large number of samples. However, as with any technology, it must be optimized and reproducible for one to have confidence in the results. Using a classical statistical method called the fractional factorial design of experiments, we assessed the effects of 11 different experimental factors. We also developed several metrics that reflect trace quality and reproducibility. These were used to measure the effect of each individual factor, and the interactions between factors, to determine optimal factor settings and thus ultimately produce the best possible traces. Significant improvements to output traces were seen by simultaneously altering several parameters, either in the sample preparation procedure or during the matrix preparation and application procedure. This has led to the implementation of an improved method that gives a better quality, reproducible, and robust output.