Previous phenological analyses of GBIF records ignored data biases and produced misleading results. Here we demonstrate how data sufficiency and methodology affect inferred latitudinal patterns in lepidopteran phenology, highlighting the importance of spatiotemporal data aggregation decisions. Results indicate later and/or shorter flight period phenology at higher latitudes is common.

Large occurrence datasets provide a sizable resource for ecological analyses, but have substantial limitations. Phenological analyses in Fric et al. (2020) were misleading due to inadequate curation and improper statistics. Our reanalysis of 22 univoltine species with sufficient data for independent analysis found substantive differences in macroscale phenological patterns.

Large occurrence datasets provide a sizable resource for ecological analyses, but have substantial limitations. Phenological analyses in Fric et al. (2020) are flawed due to insufficient data and improper analysis. Our reanalysis of 22 univoltine species with sufficient data (from 100 original species) found substantive differences in macroscale phenological patterns.