Action thresholds, based on the percentage of plants infested, for the lepidopteran pest complex in fresh-market cabbage Brassica oleracea variety capitata were evaluated in 1996 and 1997 in southern Minnesota. Three lepidopteran pests are common in Minnesota, including the imported cabbageworm, Pieris (=Artogeia) rapae (L.), diamondback moth, Plutella xylostella (L.), and the cabbage looper, Trichoplusia ni (Hübner). Most of the thresholds tested included all three pests. However, because T. ni is often the most consistent and damaging pest in Minnesota, two thresholds were based solely on the percentage of plants infested with T. ni eggs and larvae. Action thresholds were also evaluated for their compatibility with a recently labeled biologically based insecticide, spinosad, and a conventional pyrethroid, permethrin. Although all three lepidopteran pests were present in both years of the study, P. rapae provided most of the pest pressure in 1996, and T. ni was most abundant in 1997. Compared with the 0% larval infestation treatment (approximately weekly sprays from early heading to harvest), all action thresholds resulted in less insecticide use (17–80%), while maintaining high levels of marketability. Despite variable pest pressure between years, one of the thresholds based solely on T. ni (10% of plants infested with eggs or larvae) performed as well as each of the thresholds based on all three species combined. For both years, and compared with a weekly spray schedule from early heading to harvest (average of 5.5 sprays per year), use of the 10% T. ni egg or larval threshold resulted in an average of 36.5% (3.5 sprays) and 65% (2.0 sprays) fewer applications of spinosad and permethrin, respectively, with no significant loss in marketability. The results indicate that a variety of incidence-based action thresholds can be used to ensure the production of high-quality cabbage in the midwestern United States with only minimal applications of spinosad or permethrin.