How might we improve the usability and learnability of the TCSES 1.0 system?
Child Support System
Story:
The State of Tennessee's Child Support offices currently utilize the TCSES 1.0 system for all tasks and activities. This legacy system is known for its complexity and inconsistent usage across different office locations. The project involves a significant number of stakeholders, including 4,200 caseworkers and 32 judicial districts, and encompasses over 2,000 known tasks.
Company:
Deloitte Digital
Role:
UX Designer
Timeline:
Oct 21' to
Nov 22'
Problem:
The ask was to redesign the outdated legacy system, which had a reputation for being difficult to use and resulted in inconsistent usage across different offices. The challenge was to maintain the backend connections while enhancing the usability of the system.
Process:
1. Finding a Brand
To understand the brand direction desired by users for the application, I conducted a workshop where users prioritized adjectives to reveal shared values. Next, I had them vote on a scale for tones ranging from professional to warm styles. Lastly, I showed contrasting design styles, such as flat vs. dimensional, delicate vs. powerful, to determine visual preferences.
2. Establishing a Design Language
I established the design language by utilizing mood boards, style tiles, and a design system. Mood boards allowed me to gather inspiration and explore different visual directions. Style tiles were then created to communicate the look and feel of the design direction, allowing us to choose a direction early on. Finally, a design system was developed to ensure consistency in visual elements and interactions across the application. This approach helped to create a unified and cohesive design language throughout the project.
3. Validate with Users
To ensure that the design direction was on track, I used low-fidelity wireframes to gather feedback and validate the system's flow before creating the high-fidelity screens. This approach helped to avoid time and resource waste in case significant changes were needed.
4. Strategize Research Plan
To understand all of the existing user flows, I had to strategize our discovery approach. With over 2,000 known tasks, we prioritized based on frequency of use within the current application. Once I had prioritized, I used sticky notes to group tasks based on themes to consolidate similar processes.
5. Deliver High-Fidelity Designs
As I moved into the high-fidelity wireframes, I used the insights gathered from the brand and design language portion to ensure that the final designs were in line with user preferences. This saved me time by reducing the need for multiple rounds of redesigns. The resulting designs were well-received by users, who expressed their appreciation for the overall look and feel.
Lessons Learned:
Throughout the project, I gained several valuable insights, the most significant being the importance of considering bias, accessibility, and strategizing approach.
-
Firstly, I learned about the importance of accessibility during the branding phase when I had to adjust the color palette to meet AA compliance standards. One of the mood boards I created used yellow as a primary and it became apparent that meeting contrast would not be possible.
-
Secondly, I discovered the significance of bias when naming the mood boards after cities because I noticed that it influenced users' preferences.
-
Lastly, I learned that in approaching discover, it was more important to understand the groupings of the processes, rather than the specifics of each task.
Solution:
Through our dedicated efforts, the team successfully delivered the first version of the new system that showcased a consolidated view of data previously disparate in the old system. The result was a boost in the Net Promoter Score by an impressive 31 points, indicating that the new interface was well-received by users.