In this article, we present a usability experiment in which participants were asked to make intensive use of information on an intranet in order to execute job-related tasks. Participants had to work with one of two versions of an intranet: one with an organization-based hyperlink structure, and one with a task-based hyperlink structure. Efficiency and effectiveness were measured in terms of execution time and task accuracy, respectively. After the task execution, participants were asked to evaluate the task as well as the intranet. The results show that participants perform more efficiently with the organization-based structure, which is probably due to their familiarity with this structure. A post hoc analysis revealed, however, a learning effect in the task condition, which suggests that once users are acquainted with it, a task structure is at least as efficient.

Some research results suggest that declarative information in a software manual is of little value to a user. However, most research methods are aimed at measuring specific information effects. They hardly ever yield data about what users select and use if they have a choice. Also, the concepts of procedural and declarative have not been operationalized consistently. Finally, one type of user has been the main focus of investigation so far: the tutorial user. In an attempt to specifically investigate the selection and use of procedural and declarative information, a new approach is described. In an experiment based on this new approach, special attention was paid to the operationalization of the information types and three different user types were included. The results show that users use more declarative information than is often assumed.