To assist in the management of plant breeding programs, and particularly in managing change, the detailed cost structure and the nature of the cost function facing plant breeders need to be known. In this paper, an attempt is made to define a cost function for a representative wheat breeding program. The improved understanding of the cost function enables the implications for the potential uses of new breeding technologies such as genetic markers or physiological measurements to be explored to increase the efficiency of breeding programs. It also provides a basis for assessing the extent to which costs vary with the size of a breeding program.