How to use Web Analytics for Improving Web Applications Overview

As soon as a Web application has been deployed into the production environment, typically by the end of its development, some or all of its features are available to the users. By observing and analyzing the application usage over a certain period of time, it is possible to extract users ’ behavior patterns. Using these patterns to modify the application increases the chances to best meet users ’ expectations and thus improve their satisfaction. In this IBM ® Redpaper, we first introduce Web analytics: the context, the main concepts, and the common characteristics of the available tools. Then we explain how to exploit these tools to improve various points in the development and the quality assurance of a Web application, namely client compatibility, user navigation, robustness, bandwidth sizing, development, load-test plans, maintenance, and fat-client extensions. Finally, we describe a few problems and give some insights for solutions. This paper focuses on J2EE ™ Web applications, because J2EE is the programming model implemented in IBM WebSphere ® Application Server. Yet, the methods presented can be applied to any Web application technology, suc