Environmental testing of poultry premises after an outbreak of an infectious disease like avian influenza (AI) or Newcastle disease, is essential to promptly verify virus-free status and subsequently return to normal operations. In an attempt to establish an optimized sampling protocol a laboratory study simulating an AI virus contaminated poultry house with wire layer cages was conducted. Three sample collection devices, pre-moistened cotton gauze, dry cotton gauze, and a foam swab, were evaluated with each of four sample locations within a cage and when sampling all four locations with one device. Virus was detected with quantitative real-time RT-PCR utilizing a standard curve of a quantified homologous isolate of AI virus to determine titer equivalents of virus. The pre-moistened gauze detected the most virus RNA (100% positive, geometric mean titer [GMT): 3.2 log10 50% embryo infectious doses [EID50 ] equivalents per 25 cm2 ) in all four sample locations compared to dry gauze (93% positive, GMT: 2.6 EID50 equivalents per 25 cm2 ) and foam swabs (95% positive, GMT: 2.8 log10 EID50 equivalents per 25 cm2 ). The highest viral RNA loads were observed from the cage floor, and when the four locations were sampled with the same device. Overall, the pre-moistened gauze performed the best, and sampling multiple locations within a cage with the same device would likely optimize detection of residual virus.