The potential for retrieving cloud amount data from satellite-derived radiation budget archives is investigated. A predictive relationship between cloud cover and satellite-derived system albedo is established and used to produce a new oceanic cloud climatology. In view of the spatially and temporally discontinuous nature of cloud data and the diverse requirements of the three major users (i.e. scientists concerned with weather forecasting, environmental remote sensing and climate modelling), the following recommendations are made: (I) Radiance data, not cloud data, should be archived directly from satellite observations. (2) The highest spatial and temporal resolutions available should be retained in the archive. (3) Detailed comparative studies between surface and satellite assessments of cloud climatologies should be undertaken.