Assessing the Role and Value of Interactivity in Visual Business Intelligence

The development and success of many organizations to a large extent depend on how well and how timely they manage to gain insight into collected data. Business Intelligence (BI) provides methods for analyzing business-critical information and supports decision-making processes. But the ever growing amounts of data and information clearly overwhelm traditional manual methods of data analysis such as spreadsheets, standard reporting or simple diagrams. One possibility to face this problem is Information Visualization (InfoVis). By utilizing the powerful human perceptual system that is extremely efficient in processing visual input, visualization, for example, can help to make sense of data, explore complex information spaces or spot patterns and relationships within the data. Interactivity or the possibility to engage in an active discourse with the representation lies at the core of Information Visualization. However, empirical evidence and detailed approaches to model the concept of interactivity are largely missing in research today. This thesis is a first step towards bridging this gap and approaching the concept of interactivity more in depth. In order to do so, a twofold approach was chosen. First, the concept of interactivity was investigated from a theoretical point of view. In particular, cognitive theories and models were analyzed in order to determine their ability to explain and predict the value and role of interactivity in visual methods. Second, a qualitative empirical study was conducted among six IT-managers working in the field of Business Intelligence to assess the current practice at work concerning visual methods and interactivity. The main results are that cognitive theories and models do not explicitly account for the role of interactivity but rather concentrate on modeling the elements of a cognitive system and explaining how these entities interact with each other to achieve a certain goal. How different types of interactivity influence these processes is usually not part of the theories, and the effect of interactivity on cognitive processes was hardly ever mentioned. Apart from that, the empirical study showed that interactive visual methods in the area of Business Intelligence aren’t used very often. One of the main reasons reported is the fact that visualization is still two steps ahead and at the moment mostly more basic problems concerning data gathering, data modeling and data quality prevail. Moreover, most users are used to work with numbers and tables and are not aware of the possibilities in terms of visualization. However, the interviewed IT-managers acknowledged that using more interactive visual methods in Business Intelligence would be beneficial for users.