Taking a D-Tour

A New Path to Dashboard Literacy

Niklas Elmqvist
3 min read4 days ago
Authoring and disseminating interactive dashboard tours. Our workflow for crafting semi-automated onboarding experiences while preserving user agency. The content in a given visualization dashboard is extracted and converted into (a) a component graph and arranged into (b) an interactive dashboard tour, which is then (c) shared with the end-users. (From our IEEE VIS 2024 paper.)

Have you ever found yourself staring at a data dashboard, unsure where to begin? It’s a common experience — confronting a screen filled with charts, graphs, and numbers that seem to speak their own language. But what if there was a way to make these dashboards more approachable from the start?

Our team of researchers from Johannes Kepler University in Linz and TU Wien in Austria, and Aarhus University in Denmark has developed a novel approach to address this challenge. We call it “D-Tour”, which stands for interactive dashboard tours. Much like a detour can lead you on an unexpected but enlightening path there and back again, D-Tour guides users through the landscape of data visualizations, allowing for both structured learning and free exploration.

Bethesda’s Elder Scrolls: Skyrim is one of the most well-known open-world video games.

The idea behind D-Tour draws inspiration from an unexpected source: open-world video games. In these games, players have a main storyline to follow but can also explore side quests at their own pace. We’ve applied this idea to dashboard onboarding, creating an experience that adapts to users’ needs and expertise levels. In D-Tour, instead of facing all the information at once, you’re presented with an interactive guide. You can choose to follow a predefined path, learning about each element step-by-step, or you can chart your own course, focusing on the aspects that interest you most.

Typical open-world video game storyline, with main quests (center) and side quests.

Consider a sales dashboard as an example. With D-Tour, you might begin by understanding the dashboard’s overall purpose. From there, you could explore specific charts — perhaps diving into monthly sales trends or examining the breakdown by product category. The journey through the data is yours to control.

We’ve developed a tool called the D-Tour Prototype that simplifies the process of crafting these interactive experiences. Dashboard authors can easily extract information about their visualizations, arrange it into a coherent narrative, and share it with their team. Readers will then be able to adapt the dashboard tour to their expertise. An experienced data analyst might quickly move through the basics, while a newcomer to the company can take more time to understand each element in detail. It’s akin to having a flexible tour guide who adjusts the information based on your familiarity with the subject.

D-Tour Prototype’s Authoring Mode. Authors pick (a) automatically extracted visualization categories, General, Insight, or Interaction, from the Content Extraction View and drag them to the Content Arrangement View, where they (b) arrange them, (b.1) thus crafting a tour and (b.2) adding explanations to the tour content. In the Dissemination View they (c) test changes before disseminating them. A selection of the Column Chart General in the Content Extraction View is shown which is highlighted in the Content Arrangement View and in the Dissemination View. Its associated content can be seen in (b.2)

We tested our system with real users in a manufacturing company, and the results were encouraging. Authors who typically invested significant time in creating onboarding materials found they could produce engaging, interactive tours more efficiently than using standard tools. Users particularly appreciated the ability to learn at their own pace while interacting directly with the dashboard. More details on OSF.

While we’re excited about the potential of D-Tour, we recognize there’s still room for improvement. Future iterations could potentially update automatically when dashboards change or incorporate voice navigation. The possibilities for enhancing the onboarding experience are numerous.

In an era where data-driven decision-making is increasingly important, understanding our dashboards isn’t just beneficial — it’s vital. With D-Tour, we’re aiming to make the journey through data visualization more accessible and useful for everyone involved.

We’ll be presenting our research on D-Tour at the IEEE VIS 2024 conference from October 13 to 18, 2024. We look forward to sharing our findings and engaging with the visualization community to further refine and expand on this approach to dashboard onboarding.

Citation

  • Dhanoa, V., Hinterreiter, A., Fediuk, V., Elmqvist, N., Gröller, E., & Streit, M. (2024). D-Tour: Semi-Automatic Generation of Interactive Guided Tours for Visualization Dashboard Onboarding. To appear in IEEE Transactions on Visualization and Computer Graphics (Proceedings of IEEE VIS 2024). [PDF] [Website]

--

--

Niklas Elmqvist

Professor in visualization and human-computer interaction at Aarhus University in Aarhus, Denmark.