The family planning program in the Matlab District of Bangladesh has been described in unique detail for more than 25 years and is regarded as a model for equally poor parts of the world. Its experience has been reported as showing the ineffectiveness of contraceptive saturation approaches and the prime importance of program management and especially of the selection of a special type of family planning household visitor, criteria that render family planning programs relatively expensive. This reanalysis of the Matlab experience suggests that there is inadequate evidence from which to judge the record of the saturation experiment and of family planning workers from less highly selected backgrounds. It is also argued here that the role of contraceptive choice and of access to different types of contraceptives, especially injectables, delivered to the door in this society of secluded women has been underestimated, and that too little importance has been attributed to demand in contrast to supply. While it is agreed that the Matlab demonstration has been of central importance in showing that fertility can be reduced in Bangladesh, it is argued that many developing countries can draw on this experience to provide less costly family planning programs with less emphasis on the managerial, top-down approach.