Management of grape powdery mildew (Erysiphe necator) and other polycyclic diseases often relies on calendar‐based pesticide application schedules that assume the presence of inoculum. An inexpensive, loop‐mediated isothermal amplification (LAMP) assay was designed to quickly detect airborne inoculum of E. necator to determine when to initiate a fungicide application programme. Field efficacy was tested in 2010 and 2011 in several commercial and research vineyards in the Willamette Valley of Oregon from pre‐bud break to véraison. In each vineyard, three impaction spore traps were placed adjacent to the trunk. One trap was maintained and used by the grower to conduct the LAMP assay (G‐LAMP) on‐site and the other two traps were used for laboratory‐conducted LAMP (L‐LAMP) and quantitative PCR assay (qPCR). Using the qPCR as a gold standard, L‐LAMP was comparable with qPCR in both years, and G‐LAMP was comparable to qPCR in 2011. Latent class analysis indicated that qPCR had a true positive proportion of 98% in 2010 and 89% in 2011 and true negative proportion of 96% in 2010 and 64% in 2011. An average of 3·3 fewer fungicide applications were used when they were initiated based on spore detection relative to the grower standard practice. There were no significant differences in berry or leaf incidence between plots with fungicides initiated at detection or grower standard practice plots, suggesting that growers using LAMP to initiate fungicide applications can use fewer fungicide applications to manage powdery mildew compared to standard practices.