Collisions with building windows are thought to be a significant source of mortality for urban birds. Past studies on bird-window collisions have used a variety of survey methods to estimate the magnitude of mortality, and few have constructed methods in light of the biases related to scavengers and worker abilities that lead to imperfect detection of carcasses. Adoption of a systematic carcass survey protocol in future studies would reduce site-specific biases and among-site survey variability, which in turn would improve the accuracy of mortality estimates at all scales. We present here a standardized carcass survey protocol that serves two basic functions: (a) it is simple and inexpensive to implement, and (b) it accounts for the removal of carcasses by scavengers and detection of carcasses by field workers. Consistent with these goals, we added a variety of approaches with particular aspects of the protocol for researchers to chooose from depending on the goals of their study. The following topics are addressed in the protocol: 1. Preparing for surveys, 2. Supplies, 3. Frequency of surveys throughout the study, 4. When during the day to complete surveys, 5. Field worker behavior during surveys, 6. The pre-survey carcass 'clean-up', 7. How to conduct carcass surveys, 8. Carcass collection and containment, 9. Identifying species of bird carcasses, 10. Data collection and management, and 11. Duration of carcass surveys.