We present an experimental study of a flock passing through a narrow door. Video monitoring of daily routines in a farm has enabled us to collect a sizable amount of data. By measuring the time lapse between the passage of consecutive animals, some features of the flow regime can be assessed. A quantitative definition of clogging is demonstrated based on the passage time statistics. These display broad tails, which can be fitted by power laws with a relatively large exponent. On the other hand, the distribution of burst sizes robustly evidences exponential behavior. Finally, borrowing concepts from granular physics and statistical mechanics, we evaluate the effect of increasing the door size and the performance of an obstacle placed in front of it. The success of these techniques opens new possibilities regarding their eventual extension to the management of human crowds.