Studies of climate variability and change nearly always require observations of area- and time-averaged precipitation. Most hydrological studies require precipitation data suitable for studying basic-scale behavior. While precipitation is observed in various ways by a wide variety of observing systems, none of them produce precisely what is needed for climatological and hydrological studies. During the past 15 years or so, a number of efforts have been made to combine observations from rain gauges, estimated precipitation from satellite observations and radars, and other information to produce time series of gridded arrays of precipitation for the globe and for continents. The Global Precipitation Climatology Project has created time series of monthly and 5-day precipitation for the globe beginning in 1979 at 2.5° scale, and higher resolution products have been available beginning in the late 1990s. In this paper, we will discuss the issues involved in combining disparate data sources to produce optimal analyses of precipitation and describe in some detail the creation of high time-space resolution analyses for Africa and South Asia. The challenges involved in developing a precipitation analysis for North America that is capable of accurately depicting the detailed spatial and temporal variability of precipitation from a wide variety of souces, including tropical and subtropical convective activity as well as precipitation from mid-latitude synoptic systems, will be presented.