Selecting the right visualization of indicators and measures – dashboard selection model

Abstract

Background: Contemporary software development organizations utilize multiple channels to disseminate information about their indicators, measures, trends and predictions. Selecting these channels is usually done based on the availability of the visualization technology and a set of requirements elicited from stakeholders at the company. Eliciting these kind of requirements can be labor-intensive and time-consuming. Goal: The objective of this research is to develop a method for selecting which dashboard should be used. As the set of dissemination patterns of measures in modern organizations is limited, this method should be able to identify the needs of visualizations at the company and match them to the dissemination patterns and their supporting technology. Method: The research method applied is action research conducted at Volvo Car Group. The action research is conducted as part of a project redesigning a large project status reporting tool and has been designed to quantify the requirements elicited from the stakeholders of the system. Results: The results is the dashboard selection model which consists of seven dimensions - type of reporting, data acquisition method, type of stakeholders, method of delivery, frequency of updates, aim of the information, and length of data processing (flow). Conclusions: The conclusions show that using this model leads to a rapid identification of the best visualization method for measurement data, which has a cost-saving impact on measurement programs and effect-maximizing impact on the companies.