Examination of programmes of experimental work involving live animals indicates scope for worthwhile reduction in animal usage by identifying more clearly the objectives of the different stages of a programme. Having a defined first objective for new or transferred methodology as "to optimise the methods", rather than an exploratory "try it with a few animals and modify" approach, is more likely to obtain the conditions for largest divergence between the effect to be measured and the background variability. Where several experiments will use the same methodology, even small improvements in this "signal to noise" ratio may be worthwhile, as the reduction is cumulative. Subdivision to more precise objectives helps concentrate the animal usage. For example, separating "to determine the time of a distinct peak" from "to determine the size of the peak" avoids using the numbers needed to achieve the latter for every time point at which the peak might occur. Similarly, setting a series of objectives for a programme that lead to clear decision points allows setting of clear criteria for achievement at each stage, as well as for the design and analysis to be the most efficient. This paper gives illustrations of the applicability of these arguments and estimates of potential reduction.