Saliva has been championed as a diagnostic fluid of the future. Much of the attention that saliva receives as a biological specimen is due to the perception that the nature of sample collection is quick, uncomplicated, and non-invasive. In most cases, this perception matches reality; however, in some special circumstances and populations collecting saliva can be unexpectedly difficult, time consuming, and may not yield sufficient sample volume for assay. In this report, we review the nature and circumstances surrounding some of these problems in the context of developmental science and then present alternatives that can be used by investigators to improve the next generation of studies. We expect our findings will ease the burden on research participants and assistants, reduce the rate of missing values in salivary data sets, and increase the probability that salivary biomarkers will continue to be successfully integrated into developmental and behaviorally-oriented research.