AbstractThe purpose of this article is to explore the nature of the “output” of a university and of the inputs to it, in order to see how far measures of productivity or efficiency can be devised. It is assumed to be desirable to use the resources of the institution, necessarily limited, as efficiently as possible in the pursuit of its objects. But there are logical difficulties about any overall measure of efficiency: a university is not like a coal mine. However, useful results can be obtained from a partial approach, and a number of suggestions are offered. A central difficulty is the lack of a sufficient understanding of the nature of the work‐load which is most likely to make a person of high intellectual ability effective. It is suggested that a variety of tasks is essential: and this has implications for the use of ancillary staff. A number of sub‐sections of university activity can be made more productive provided that minimum standards of performance can be specified. This is occasionally easy, but the specification of performance for (e.g.) central administration is a difficult matter. It is suggested that over‐enthusiasm for the better use of buildings and equipment may not always be wise, but that common‐sense and operational research can nevertheless lead to significant savings in the use of capita