Problem Statement:

Gary Community Ventures aimed to redesign their benefits screener web app to better serve users—particularly those with accessibility needs—by reducing drop-off rates, clarifying navigation, and improving overall usability. The existing interface failed to communicate value clearly, lacked intuitive filtering, and did not meet essential accessibility guidelines

Key Challenges:

• Unclear Information Architecture: Users struggled to understand where they were in the process, which led to high drop-off rates and poor task completion​.

• Limited Accessibility: The experience was inequitable for users with disabilities or assistive technology needs, creating barriers to accessing critical resources.

• Filtering & Navigation Issues: Users couldn’t easily filter benefits by eligibility, citizenship status, or immediate vs. long-term needs, which caused confusion and limited the tool’s utility​.

• Missing Support Pathways: Assistance options (e.g., “get help” or case manager referrals) were buried or unintuitive, and users were unsure where to go for next steps​

Solutions:

• Co-Design & UX Audit: Conducted stakeholder working sessions and a UX audit to align on user goals and define key areas for improvement.

• Iterative Testing: Completed two rounds of in-person and virtual user testing with over 10 participants, directly informing changes to layout, terminology, and filtering logic​​.

• Redesigned Filtering & Results Pages: Introduced eligibility-based filtering, clearer “save my results” options (email, text, print), and removed unnecessary steps like "view results" to streamline the experience.

• Accessibility Enhancements: Improved structure, color contrast, and labeling throughout the interface, and added clarifying info icons and simplified content to improve usability for all users​.

• Clearer Pathways to Action: Added direct links to application assistance and emergency aid pages, surfaced “get help” options more visibly, and introduced benefit banners to guide users toward next steps.