Both herbarium specimens and citizen-science observations have become increasingly accessible but underutilized sources of data for phenology research. We developed and tested a novel technique that uses high-quality citizen-science observations to provide an independent validation of phenology models derived from herbarium specimens. Using this approach, we were able to evaluate the strengths and weaknesses of the herbarium data set. We found that our models were valid for understanding the relative shifts in the phenology of the focal species across a large geographic area but needed further calibration to provide accurate predictions for specific dates and locations. As herbarium specimens become increasingly digitized and citizen science becomes more frequently used as a research methodology, the potential application of the technique we present here will grow.