NATIVE MOBILE | MASTER'S THESIS 2024
Designing ShelterLink: A Real-Time Emergency Housing Ecosystem
A 0-1 mobile app eliminating data latency to streamline real-time shelter bookings for managers & volunteers and help the unhoused.
✿ Context: Master's Thesis HCI (CSU East Bay - Interaction Design)
✿ Duration: Aug 2024 – Dec 2024
✿ My Role: UX Researcher, Product Designer & Front-End Developer
✿ Platforms: Native iOS & Android
✿ Scope: Generative Research, Dual-User Journey Mapping, UI/UX Design, and Front-End Development (React Native + Firebase).
✿ Tools & Workflow:
✿ Ai Workflow:
Overview
✿ The Product: A 0-1 native mobile ecosystem designed to bridge the communication gap in emergency housing.
✿ The Solution: An end-to-end app that syncs shelter capacity in real-time.
✿ The Impact: Transforms a chaotic, manual booking process into a seamless digital tool.
Problem statement
✿ Analog Systems: Locating emergency beds relies entirely on outdated paper lists and endless phone calls.
✿ Data Latency: Fragmented, delayed information makes it impossible to track real-time bed availability.
✿ Severe Consequences: This friction results in wasted travel for the unhoused and massive operational bottlenecks for shelter staff.
❌ The Problem (Analog & Fragmented)
Data Latency: Information relies on endless phone calls and paper lists, making it instantly outdated.
Wasted Travel: Unhoused individuals walk miles only to find shelters are already at maximum capacity.
Operational Bottlenecks: Shelter staff are completely overwhelmed trying to track bed inventory on whiteboards.
✅ The Solution (Digital & Real-Time)
Live Sync: A centralized Firebase database updates bed capacity instantly across all devices.
Smart Routing: Google Maps API integration connects users directly to confirmed open beds nearby.
Manager Dashboard: A dedicated, one-tap interface allowing staff to update bed counts in seconds.
The Core Goal
✿ Shift the cognitive and logistical burden of finding a bed from the unhoused to equipped volunteers using a dual-sided, real-time ecosystem.
The Competitive Gap
✿ Existing apps are just static directories.
✿ Because they lack live API integrations, they force vulnerable users right back into a frustrating cycle of manual phone calls to confirm space.
A quick explanation of the problem statement and my research strategy.
Generative Research & Methodology
Key insights from interviews:
✿ Outdated Information: Volunteers and managers face delays due to missing or outdated bed data.
✿ Communication Hurdles: Multiple calls or visits are required to confirm bed availability, and assistance is slow.
✿ New Volunteer Challenges: Clear guidance is needed for inexperienced volunteers.
✿ Data Transparency: Managers want a real-time, centralized platform for updates.
✿ Trust Building: Reliable data is critical for volunteers to assist effectively.
%20copy.png)
Generative Research (N=9)
Working alongside local organizations like South Hayward Parish, I deployed a mixed-methods approach across 9 stakeholders (volunteers, managers, and advocates), uncovering two core insights:
✿ The Volunteer Friction: The Volunteer Friction: Analog bed-searching (endless calls and walk-ins) causes severe delays.
✿ The Manager Needs: Urgent demand for a real-time, synchronized system to broadcast bed availability.
.png)
.jpg)
.jpg)
Core Research Objectives
✿ The Seekers: Identify the major pain points unhoused individuals face when accessing shelters.
✿ The Operators: Uncover the systemic challenges slowing down shelter staff and managers.
✿ The Ecosystem: Explore opportunities to seamlessly integrate food and extended support services.
In person Key Interview Findings
✿ Wasted Travel: Unhoused individuals are consistently misdirected by inaccurate, delayed bed counts.
✿ Operational Bottlenecks: Shelter staff are overwhelmed by inefficient, manual tracking systems.
✿ Fragmented Coordination: Volunteers lack the real-time updates and scheduling clarity needed to assist effectively.
Stakeholder management mapping
The Product Lifecycle
Ideate
Transform core user friction into a clear MVP strategy.
Design
Prototype the dual-sided ecosystem to align with user needs.
Develop
Code and deploy the final screens into a live application.
Validate
Measure success by tracking how fast volunteers can book beds and how easily admins can manage them.
Ideation
Sketches, rounds 1 & 2
story board
Mentoring session with Marina Terteryan
Tackling challenges
Archetypes and Their Needs
Need: Quick access to bed information, easy-to-use interface, and communication with shelters.
Task Journey Map
volunteers' map
✿ Pain Point: Lack of real-time updates on bed availability.
✿ Opportunity: Introduce instant notifications and booking.
shelter manger/staff map
✿ Pain Point: Managing multiple requests manually
✿ Opportunity: Create an organization. real-time request management system.
Information Architecture
Code Snippets
Shelterlink Techstack
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
✿
The Build
Prototype and Iterations
Visual Identity
ShelterLink pairs the trust-building, reliable tones of a blue and white palette with highly accessible typography (Poppins) to reduce cognitive load and evoke security during high-stress emergency bookings.
