The TGI, of course, is not the first single-source media product survey to be set up on a continuous basis. The TEM system in Sweden, and the Brand Rating Index and Simmons surveys in the USA, also work elsewhere on data marriage, are all extremely relevant and are no doubt familiar to many of those attending this Congress. The writer, however, has been closely involved in the organisation and operation of the TGI from the beginning and hopes that his experience, limited though it is to Britain and to one survey, will be of interest to those working in this field. It is not the purpose of this paper to explain in detail how the TGI works. In brief, a nationally representative sample of adults is contacted, the interviewing being spread evenly through the year, a certain amount of classification information is collected from them, and they are asked to complete themselves a long questionnaire, of which there are different versions for men, housewives, and other adult women. About sixty percent of the contacts return questionnaires which are acceptable for analysis. The results included in this paper relate to October 1968 - March 1969.