This paper briefly reviews the measurement of precipitation by radar, discusses factors affecting the accuracy of such measurements, and outlines how such factors may be dealt with to improve the quality of precipitation measurements by radar for the purposes of the Baltic Sea Experiment (BALTEX). Precipitation products from the BALTEX Radar Network (BALTRAD) are then briefly presented, along with descriptions of how their qualities are improved, as are some new results on their accuracies. Intelligent compositing of data from a heterogeneous network, combined with innovative quality control, is shown to give high quality high resolution information for monitoring relative precipitation variability simultaneously over land and sea in both time and space. Gauge adjustment of radar-derived accumulated precipitation is shown to efficiently minimize the radar data's bias with increasing distance, thus yielding quantitatively useful datasets for application by the BALTEX community.