Creating data visualisations for dashboards using participatory design and low-fidelity prototyping

Abstract

As the business world becomes more data-driven, it is becoming increasingly important to gain insights into organisational data to gain understanding and thus gain competitive advantages. Business intelligence is the term used for the methods and tools used by organisations to gain this understanding. One of these methods is data visualisation, which presents insights into the data in a supposedly easy-to-digest format to the information consumers. A common tool used for data visualisation in organisations is a dashboard, which is a collection of informative visualised data pieces. There are, however, some problems with data visualisation and dashboarding within organisations. Visualising data is a complex task, as the creator of the visualisation needs an understanding of how the human visual perception system works. There is also the matter of the visualisation being open to interpretation, and some users might not understand the visualisation as intended. In addition, there are issues directly related to dashboards. Creating dashboards can seem like an intimidating task, and some users might feel that dashboards attempt to oversimplify the intricacies of their organisation. Adding to these problems is the perception from management that their inputs into deciding which metrics to display on the dashboard was not requested. This study proposes making use of participatory design and low-fidelity prototyping to get around these problems. Making use of participatory design will help give a voice to the users, as they are active in the design of the dashboard. Low-fidelity prototyping is a low-cost and practical way to create the prototypes, as it makes use of inexpensive items, can easily be discarded when a mistake has been made, requires less time than a high-fidelity prototype, and stimulates creativity. By making use of design science and case study research, participants were tasked with creating a data visualisation based on a company problem. A single design iteration was used. The observations, and post-session interviews were used to create a method, the artefact of the design science research. This model can now be tested and refined further.