This paper will discuss a practice of utilizing Web logs and usage statistics to maximize library resource and improve services at the University of Colorado Health Sciences Center Library. Like many of its peers, in 2004 the Library switched most of its print journals to online subscription. In this transformational process, the Library was faced with unique issues related to digital resources and services. Questions concerning reliable usage measurements, proficient access management and control, adequacy of user licenses, robust system performance and IT infrastructure, strong compliance of copyright laws and contractual agreements lead to the development of a decision-support mechanism based on system-based usage logs, combined with data collected from qualitative measurements.

By systematically and periodically mining the data collected from library servers, click-through tracking script, and electronic resource vendors, usage patterns are identified and studied. Statistical reports provide evidence for collection development, resource allocation, budget and funding requests, system performance, and IT capacity planning. In addition, log analysis can also reveal unusual activities and potential threats to network security. From the log entries, we have observed our Website was cloaked by massive and exploitative Web crawlers from Google and other search engines. At times, we are alerted by attempts of unauthorized access and excessive downloading. The warnings lead us to take corrective actions before problems further escalate. Such practice further fosters a culture of evidence–based assessment on services and resources in the Library.

In a networked academy, libraries, together with their parent organizations, are striving to serve their less-visible, but more diverse customers with ever-growing virtual resources. Methodically monitoring and studying network traffic and usage statistics lay a foundation to better serve our users with adequate digital resources and reliable IT infrastructure.