Where are residential customers using their water? How can subtle effects of residential water conservation programs be accurately measured against highly variable background water use? These questions are of critical importance to water providers and planners who are routinely called upon to project water demands, system reliability and determine the requirements for massive capital projects needed to meet these demands. A new data collection technique developed in Boulder, Colorado, makes it possible for water providers to not only precisely measure residential water use, but also to quantify the actual water savings effected by any conservation program. This paper describes a pilot study conducted in a suburban community in Boulder that uses a single data logger attached to residential water meters to produce disaggregated data. Nine distinct residential water use categories were identified for this study: toilets, showers, bathtubs, dish washers, clothes washers, irrigation, faucets, leaks, and swimming pools. A computerized customer information system (CIS) was also created to aid in the selection of the study group.