Estimating exclusion limits in prospective studies of searches

We discuss the procedures used in studies of a prospective nature to estimate the minimum sample size (or luminosity) needed from a statistical point of view to put exclusion limits to a given signal. We show how such limits should be computed in the most general way and compare the results to other more restricted approaches. This allows us to point out some ambiguities in the way prospective estimation of exclusion limits is sometimes carried out.