The lack of effective surrogates or biomarkers for biologic and disease state is a major hindrance to effective cancer care. Many laboratories have been using mass spectrometry as the means to discover new biomarkers. The success of these efforts relies on effective collaboration between clinicians and mass spectroscopists. This review describes many of the common proteomic experimental methodologies and evaluates the current state of proteomics-based biomarker discovery--starting from sample collection through sample processing and ultimately sample measurement. Guidelines are given for the evaluation of experimental design and for the gauging of result significance. Particular emphasis is placed on those aspects strongly associated with reproducibility from both clinical and laboratory perspectives.