Statistical systems are a consequence of evolution and the level of data integration that is achieved is often an indication of the degree of system development. Almost all national statistical agencies in the world integrate administrative data with their survey and census information to some degree in order to complement, supplement or replace survey information or to assist with frame maintenance. This paper reconstructs the development and evolution of the Canadian agricultural statistical system as it relates to the expanding and increasingly important role of administrative data. The degree to which administrative data are integrated depends on a number of factors, the most important being: (1) the degree of maturity of the country's statistical system, (2) the quality and the amount of information available from the government's administrative and regulatory programs, (3) well-trained and experienced staff, (4) funding, and (5) co-operation among government agencies. Most countries appear to have gone through an evolutionary process in establishing their agricultural statistical system and most systems are developed with an internal capacity for renewal and adjustment. This allows them to respond to changing conditions and needs and to remain relevant. The actual route that is followed, however, is highly dependent on the amount of resources available for the program, the availability of experienced professionals to develop and maintain the system and the statistical toolbox that they are able to use.