Advances in biochemical genetics have resulted in the ability to detect carriers of some recessively inherited diseases. Instigation of control programmes based on this ability will depend, at least partly, on the costs and benefits to industry, individual breeders or members of a breed society with a vested interest. Such programmes may involve all animals in a particular population or be more selective. These factors are analysed and discussed using the bovine mannosidosis programme in New Zealand as a model. At a heterozygote prevalence of 10%, there were benefit:cost ratios of 2.89:1 and 4.14:1 over a 20-year and infinite-time horizons, respectively. When lower prevalence figures were examined, present value costs equalled present value benefits when 5 or 6% (depending on time horizon) of the population were heterozygous. The model developed and principles discussed should prove useful in developing similar, or alternative, schemes for this or other diseases of livestock.