We propose a measure of reliability called information volatility (IV) to complement Business Intelligence tools when considering aggregated data or when observing trends. Two types of information volatility are defined: intra-cell and inter-cell. For each, two types of distributions are considered: normal and lognormal, which is often the case for time series data. The IV measures are based on similar measures found in the finance literature, since there are similarities in the types of data. In order to understand the information volatility metrics, the notion of benchmarking is introduced with three propositions: numerical benchmarking, graphical benchmarking and categorical benchmarking. The IV metric is designed and evaluated using the design science research paradigm: first, the metric is developed and then it is evaluated through the use of focus groups (including several cycles for refinement of the design). The paper concludes with our research contributions and future research directions.