Dashboard & Data Visualizations
A reporting and dashboard design that provides insights into areas where organizations may be at risk of not meeting regulatory and compliance standards.
Problem
1. Design system was not mature
At the start of this design process, our design system was still a work in progress, and we lacked any patterns in our library for data visualization components. Therefore, we had to spend a lot of time researching and conceptualizing components before really tackling the "solutioning" steps of creating the dashboard. To ensure that our components were accessible, we needed to make sure that data could be read with various visual impairments by checking contrast and attempting to find patterns we could use in lieu of color in some places, which proved to be difficult.
2. Custom components needed to be built in PowerBI
Similar to what I experienced later on when creating a reporting solution that worked hand-in-hand with our dashboard, out of the box PowerBI components did not have enough customization for our needs. I worked with engineers to come up with a set of initial data visualizations that we were happy with and met our data needs.
3. Proposed data controls were difficult to make in PowerBI
We initially hoped to include the ability to control which data points were displayed on each axis by having an expansion panel that displayed a menu where users could select a new data point, but had to hold off on this for MVP. Since this was the first project we did in PowerBI, we did not have a lot of time or experience that would have been needed to find a solution for this, so we removed this feature for now.
4. Limited color palette
Due to the early stage of our design system, we had a limited color palette for data visualization. We collaborated with the creative team to develop colors that met both brand and user needs, but creative differences and the lack of established data visualization components posed challenges. Given our time constraints, we used the initial draft of the color palette from that collaboration, ensuring sufficient contrast levels between data points. However, testing with a vision simulator tool revealed that the initial color ordering was insufficient and would need further refinement in the future to truly be accessible.
Problem Statement
How might we provide valuable, accessible data to our users at glance by incorporating a dashboard into our product using PowerBI?
Proposed Solution
For the MVP, I proposed a dashboard that included filters and various data visualizations, utilizing our available colors to ensure maximum accessibility. Due to PowerBI constraints, we had to adjust our filters to be exposed rather than housed in an expandable filter panel. Initially, our designs featured only five data components on the dashboard, but by May 2024, we were able to add two more. We also established a small set of data visualizations that now serve as foundational components of our design system and are used across the platform.
Findings & Next Steps
So far, the feedback on the dashboard has been very positive. Initially, I was concerned about making the PowerBI integration seamless, but the engineers did an excellent job translating my designs into custom components.
Next, we aim to further enhance the risk dashboard experience by incorporating AI and risk monitoring to generate recommendations that help our users stay compliant. Ideally, I would like to see automated corrective action plans created based on incident reporting data and suggested to our users.
As we include additional data points, we will need to revisit adding more visualizations and continue improving our color palette for better accessibility. My goal is to develop an in-depth color palette similar to IBM's Carbon Design System, with an established color order and variations for a specific number of data points.
Background
As part of the new compliance suite, we needed to create a dashboard for users to monitor their organization's areas of risk at a glance. This dashboard will eventually include data on policies, incidents, and regulatory information. For the MVP of the risk dashboard, I collaborated with another designer to develop initial data visualizations, iterating on them as we approached launch and as our data needs expanded.
Stakeholders
Product Managers, Software Engineers, Test Engineers
Role
Lead UX Designer