To tackle this complexity, I adopted a grounded theory approach, conducting structured and semi-structured interviews with 119 participants across 13 workgroups. We cast a wide net by talking to first responders, program administrators, and citizens, because a mental health crisis touches every layer of the city. These were not quick chats. These were 90-minute deep dives designed to uncover the personal stories hidden behind the statistics.

Our strategy centered on participatory technology development. Instead of building for the community, we engaged stakeholders as partners in the development cycle. We used frameworks such as the Transtheoretical Model of Health Behavior Change to map the user’s mindset from “unaware” to “proactive”. This approach revealed that social determinants, such as housing stability and food security, were just as critical as clinical therapy. We mapped the ecosystem and identified distinct user journeys, such as one for individuals seeking services and another for providers making referrals. We utilized journey maps as the key driver of our design decisions, ensuring we were solving for the “whole journey” rather than isolated touchpoints. We even developed specific archetypes, like “The Techie” vs. “The Browser,” to ensure our interface could flex to meet different cognitive styles and stress levels.

We resisted the urge to build everything at once. We proposed an MVP built on three strict requirements: usability/accessibility, SDoH integration, and data security. Before the full build, we deployed an “Alpha” version for testing to validate our core assumptions.

The pilot launched in March 2022, kicking off a rigorous cycle of iteration that ran through June 2022. We treated the product as a living organism. AB testing was conducted through specific user workgroups. For example, during testing of the resource listings, stakeholders rejected “Option 1,” which required an extra click to view services, preferring “Option 2,” which displayed details immediately. We listened and pivoted.

Our implementation strategy was phased. We focused first on the core functionality for individual users, such as the home page, the SDoH screener, and resource listings, before attempting the complexity of the provider dashboard and referral tracking system. Every phase was checked against user feedback, particularly regarding sensitive privacy concerns. We used tools like Maze to generate heatmaps, confirming that 76.7% of users could successfully navigate to the assessment on their first try, giving us the confidence to scale.

During our sense-making process, several pivotal themes emerged that reshaped our design criteria:

• Trust is the Gateway. Transparency is mandatory, as participants communicate the importance of service transparency in understanding who sees their data.
• It’s not just mental health; it’s whole-life health. Social determinants like housing stability and food security significantly affected access to services. Participants noted barriers when clients lacked phones or housing. We could not just list therapists. We integrated the SDoH screener to capture these broader needs and connect them to relevant resources, such as food pantries and housing support.
• Barriers are multifaceted. Participants reported barriers ranging from a lack of awareness to transportation and linguistic obstacles. The design needed to incorporate best practices for accessibility and address cultural stigma.
• The system is fragmented, focus on the journey. Current care is focused on quantity over outcomes. Providers were maintaining their own Excel sheets because they did not trust central lists. The platform had to focus on individual journeys rather than isolated system touchpoints, facilitating self-referrals and tracking progress so no one fell through the cracks.

The experience strategy needed to translate empathy into measurable system performance. We aligned our design decisions around three core tiers of success:

Goal: Operationalize Trust. We aimed to track engagement with the sensitive SDoH screener as a proxy for user trust and safety.

Goal: Close the Crisis Gap. The Challenge: Users in distress face an infinite loop of referrals and retold stories, often abandoning the search. The goal was to close the gap between identifying a need and receiving care, and to monitor metrics like “referral follow-through” and the reduction of “multi-contact” friction, where users had to call multiple places to get help.

Goal: System Integration. Providers relied on isolated personal Excel sheets, creating a fractured view of available care. We aimed to track the unification of disparate databases into a single source of truth to drive the $18M transformation initiative.