Journal citation data is valuable as a selection tool for adding new journals as well as for discontinuing subscriptions that are no longer cost-effective. This article presents and discusses an example of data extraction from a typical ISI Web of Science report. The strategy was developed following a review of the data relationships and embedded data output format. While Perl was used in the example, the method described can be implemented with most programming/scripting languages. The example demonstrates also that citation-based studies and reports can be based on large sets of extracted data rather than the typical, small samples. The value of the data is discussed using a actual decision-making scenario.