Despite great interest, investments, and efforts, the ongoing search for plasma protein biomarkers for disease so far has come up surprisingly empty-handed. Although discovery programs have revealed large numbers of biomarker candidates, the clinical utility has been validated for only a very small number of these. While this disappointing state of affairs may suggest that plasma protein biomarkers have little more to offer for diagnostics, we take the perspective that experimental conditions might not have been optimal and that analyses will be required that offer far greater sensitivity than currently available, in terms of numbers of molecules needed for unambiguous detection. Accordingly, techniques are needed to search deep and wide for protein biomarker candidates. The requirements and feasibility of such assays will be discussed.