Although the system usage construct has a central place in information systems (IS) research, there has been little discussion to date about how it should be measured. Usage is most commonly measured by self-reported questionnaire data, and, on occasion, interview reports, computer logs, or independent observation. Some researchers have found that the relationship between usage and other constructs differs depending on its method of measurement. Consequently, there is a need to determine how to obtain ‘true’ measures. The objective of this paper is to present a framework for measuring systems use. Specifically, the paper distinguishes two components of systematic variance—common methods bias and distance bias. These can lead to inaccurate measures of usage and inaccurate measures of its relationships with other constructs. The paper then presents an approach to minimizing these two sources of systematic bias and operationalizes this approach in the context of studying the relationship between usage and performance.