Research on neonicotinoids and honeybees have changed focus from direct mortality to sublethal effects. In the present study, a published honeybee model, BEEHAVE, is used to compare induced colony level impact of pesticides including direct mortality, poor brood care, disorientation, and increased handling time in oilseed rape and sunflower crops. Actual effects on individual bees will depend on exposure concentrations, but in the present study large effects were enforced. In oilseed rape, poor brood care had the largest colony impact, because it created a bottleneck for spring build-up of the workforce, and colony impact for all effect types peaked 1 mo after exposure ceased. In sunflower, the later exposure changed the response so colony impact peaked during exposure, and the bottleneck was honey store build-up. In all scenarios, good forage mitigated effects substantially. It is concluded that field studies should continue at least 1 mo after exposure to ensure detection of ecologically relevant sublethal effects. The results indicated that even if a sublethal effect is difficult to detect in the field, subsequent ecologically relevant colony level impacts would be clear if studies are continued for 1 mo after exposure. Guidance for regulatory studies recommends extended observation periods, and published field studies already use extended observation periods, so it is concluded that current methods are adequate for detecting ecologically relevant sublethal effects. Although published laboratory and semifield studies conducted under controlled exposure conditions suggest that sublethal effects may occur, published field studies with neonicotinoid seed treatments, naturally foraging bees, and extended observation periods do not report colony-level effects, suggesting that in these studies no ecologically relevant sublethal effects occurred. Environ Toxicol Chem 2017;36:831-840. © 2016 SETAC.