A successful management plan requires clear goals and a process for evaluation. Without them, managers risk operational shifts in which continuous changes disguised as improvements may have little beneficial effect. The conception, design, and operation of an electric barrier and fishway on the Pere Marquette River of Lake Michigan serve as an illustration. The Great Lakes Fishery Commission operated an electric weir to stop migration of adult sea lampreys (Petromyzon marinus) and a fishway to provide upstream passage of rainbow trout (Oncorhynchus mykiss) during 2000–2009. The weir and fishway were successful in blocking some of the annual spawning run of sea lampreys (trapped an annual average of 439 sea lampreys) and it allowed passage of rainbow trout (an annual average of 6,091). Even with success that yielded an estimated density of 0.03 adult female sea lampreys per 100 m² of larval habitat upstream of the weir, lampricide treatments continued because the weir still allowed establishment of substantial densities of larval sea lampreys. Our evaluation suggests that an in-stream barrier must approach 100% blockage of sea lampreys to eliminate large recruitment events. The failure of the weir to reduce lampricide treatments was due to an informally defined purpose and measures for success at the onset, the complexity of electric weir systems (and the operational problems created by such intricacy), and lack of recognition of the reduction in larval sea lamprey recruitment needed to succeed. Control of sea lampreys in the Pere Marquette River could have benefited from an adaptive management approach.