<p>Invasive social wasps are a growing concern in New Zealand, posing both ecological and economic threats. However, gathering large datasets on their phenology and seasonal dynamics can be challenging. In this study, the caste-specific seasonal activity and spatial distribution of three invasive social wasp species (<i>Polistes dominula</i>, <i>Vespula germanica</i>, and <i>Vespula vulgaris</i>) in New Zealand were investigated using online citizen science data. The original dataset, sourced from the Global Biodiversity Information Facility (GBIF), contained 3554 observations spanning from 1945 to 2024. Following expert validation, which focused on improving data quality and extracting caste information, a refined dataset of 2028 observations was produced, from 2005 to 2024. Although citizen science data were generally accurate for species identification, expert input substantially improved their utility by providing caste information. Seasonal patterns of reproductive castes aligned with known wasp phenology, with a notable peak in male activity during autumn. Spatially, observations were concentrated in urban and lowland areas, suggesting a potential bias in citizen science data. An ecologically significant finding was the observation of 111 winter-active workers across species, on both islands, and throughout all three winter months. These workers, not confined to urban settings, could indicate potential perennial nesting. The utility of citizen science for ecological monitoring was emphasised in this study, particularly for identifying caste-specific seasonal patterns, such as autumn male activity, which could inform more targeted pest management. Additionally, the findings highlighted the need for further investigation into winter activity and its implications for wasp management.</p>