Feeding birds is a common activity throughout the world; yet, little is known about the extent of feeding gulls in urban areas. We monitored 8 parking lots in central Massachusetts, USA, during the fall and winter of 2011 to 2013 in 4 monitoring sessions to document the number of gulls present, the frequency of human–gull feeding interactions, and the effectiveness of signage and direct interaction in reducing human-provisioned food. Parking lots were divided between “education” and “no-education” lots. In education lots, we erected signs about problems caused when people feed birds and also asked people to stop feeding birds. We did not erect signs or ask people to stop feeding birds at no-education lots. We spent >1,200 hours in parking lots (range = 136 to 200 hours per parking lot), and gulls were counted every 20 minutes. We conducted >4,000 counts, and ring-billed gulls (Lorus delawarensis) accounted for 98% of all gulls. Our educational efforts were minimally effective. There were fewer feedings (P = 0.01) in education lots during one of the monitoring sessions but significantly more gulls (P = 0.008) in education lots during 2 monitoring sessions. While there was a marginal decrease (P = 0.055) in the number of feedings after no-education lots were transformed into education lots, there was no difference in gull numbers in these lots (P = 0.16). Education appears to have some influence in reducing the number of people feeding gulls, but our efforts were not able to reduce the number of human feeders or the amount of food enough to influence the number of gulls using parking lots.