This paper takes up the current discussion and disagreement among ethnomethodologists and conversation analysts concerning how conversation analysis should address questions of "social structure." It also discusses the question of whether conversation analysis can address questions concerning the organisation of "work" as developed in the "studies of work" program of ethnomethodologists. Five different types of ethnomethodological and conversation analytic studies are delineated in order to show that, altough they differ in problem selection and formulation, methodological preference and foci, they are not incompatible but complementary.