Benefits Enrollment
In 2016, the US Digital Service (USDS) wanted to study why a large portion of Americans who qualify for social services, such as SNAP and Medicaid, are not using those services. One hypothesis, was that the systems governing enrollment for these services were negatively impacting usage rates for qualifiers of these programs.
In order to increase the enrollment rate for qualified individuals, USDS spearheaded a project to research the current experience of people using this system and develop a prototype to test a possible solution. I was engaged to design and staff the research project, and lead that research team.
Goals
Observe and understand the experience of enrolling in – or supporting someone to enroll in – multiple social benefits in different areas of the US. Observe and understand the experience of those facilitating within the system.
- Build greater understanding of people who use government enrollment systems
- Understand the benefits enrollment experience from both the enrollee and the staff who run the process
- Gather data to build personas
- Develop a report for coalition building and budget approvals
- Generate a prototype to illustrate the possibilities
- Investigate enrollment business needs, pain points and successful common practices
- Map user journey through the enrollment processes
Approach
USDS partnered with 18F to design and execute the qualitative research study. As research lead for the project, I designed the study as a set of contextual inquiries and user interviews in multiple locations. Our available staff was had little to no research experience, so I selected individuals who hat the highest aptitude for the work and I created a training program for doing non-leading interviews and gathering qualitative data formatted for quick synthesis. Our team worked in pairs, trading off facilitation and note-taking roles.
- 40 applicants
- 42 assisters/community-based org staff
- 15 state and local administering agencies
To expedite synthesis of this large volume of qualitative data, we developed a realtime coding system which used descriptive and process tags. After each session and during end of day team check-ins, we aligned our coding and optimized the system. The team performed synthesis both in-person and via remote collaboration methods.
Key Insights
- The process of enrolling and re-certifying for benefits in each locality have critical failure points which rely on mechanisms which frequently fail due despite best efforts of applicants.
- Benefit programs are not connected, and each requires applicants to perfectly execute a different set of steps and qualification requirements.
- Enrolling and staying enrolled in benefits consumes a significant amount of time by any standard.
- Offices which process benefits applications are understaffed, and struggle to execute the cumbersome process required by their systems and statutes to meet demand.
- The best programs Take on the responsibility of verifying eligibility, either manually or using existing data, to reduce the burden to applicants.
- Current online experience is cold, hard to understand, and cumbersome for staff and applicants alike.
Solutions
A set of recommendations were established as guiding principles for future development of an open source system which could be adopted by municipalities to replace their old, non-compliant systems. This platform is architected to allow each municipality to set it up with their own business rules as these rules are defined independently by each state across the country.
Recommendations
- Deeply invest in increasing efficiency, continuously tracking operations data and analyzing effectiveness of their processes.
- Integrate program applications to help serve individuals and families more holistically.
- Use technology and multi-channel touch points to communicate in ways that help keep people from dropping out during enrollment or renewal.
- Empower staff to use their judgment and discretion, moving away from a ‘line worker’ model, increasing effectiveness and satisfaction for both staff and applicants.
Full Research Report