top of page
D (1).png
Image by ActionVance

Philadelphia Housing Authority 

Navigating for Everyone: A UX Research Study of the Philadelphia Housing Authority Website

Role

UX Researcher

Tools

Figma

Timeline

2 Months

Team

3 UX Researchers

Overview

This project was a collaborative UX research engagement with the Philadelphia Housing Authority, conducted as part of my graduate studies. Working with a team of three designers, I contributed equally across the full research process: developing personas, building and analyzing journey maps, running competitive analysis, and conducting information architecture testing through card sorting and tree testing. The output was a set of research backed recommendations for improving how users navigate and access housing resources on the PHA website.

Goals
  • Simplify navigation and information retrieval.

  • Improve website accessibility and inclusivity.

  • Ensure a user-centered design approach based on real user needs.

  • Align website functionality with the PHA’s mission of providing housing resources.

The problem

The Philadelphia Housing Authority website serves a wide and varied user base: residents applying for housing assistance, landlords participating in the Housing Choice Voucher program, seniors seeking accessible housing, and people in crisis needing emergency resources. For many of these users, the stakes are high and the margin for confusion is low.

Our research set out to answer four questions:

  • What are the main pain points users face when navigating the current PHA website?

  • How can the information architecture be improved to better support user tasks?

  • What accessibility features would best serve PHA's diverse user base?

  • How does the PHA website compare to competitors in usability and functionality?

What we found was a site that consistently worked against its users. Key problems included:

  • Navigation organized around PHA's internal structure rather than user tasks

  • Emergency and crisis resources buried several levels deep

  • A search function that could not handle misspellings, synonyms, or alternative phrasing

  • Inconsistent branding on external login pages that undermined user trust

  • No clear path for new applicants trying to apply for housing

Users

One of the first things we established was that the PHA website does not serve a single type of user. We developed four personas to ground our research: Arthur, a landlord exploring the Housing Choice Voucher program; James, a retired senior looking for accessible housing with limited computer experience; Jay, an emergency voucher recipient struggling to find support while his voucher neared expiration; and Amara, a single mother trying to understand what housing programs she could access.

 

The range mattered. A design decision that worked for Arthur, who approached the site like a business transaction, could completely fail James or Jay, who came with more urgent and fragile needs. Journey mapping each persona through real scenarios on the site made the gaps concrete.

Research and Methodology Insights

01. Open Card Sorting

We conducted open card sorting with 20 participants to see how they naturally grouped 50 key tasks. This revealed strong mental models around specific life-stages rather than departmental silos. Users consistently grouped "Apply for Programs" and "Waitlist Status" together, whereas the current site separated them by administrative departments.

Card Sorting.jpg
Tree Testing.jpg
02. Tree Testing Analysis

We followed up with tree testing to validate our proposed labels. Results showed a 45% increase in findability when using task-based labels over department names. The heatmaps clearly showed that "Emergency Housing" was now discoverable on the first attempt by nearly 90% of participants.

Competitive Analysis

We reviewed four comparable organizations, This gave us a baseline for what accessible, well-structured housing authority sites looked like in practice and surfaced specific features worth recommending, including embedded accessibility widgets and clearer login page conventions.

HACP_Logo_RGB_Fullname_Horizontal_FullColor_Left_Clear_scaled.png

Housing Authority of Pittsburgh

sdhc-logo.png

San Diego Housing Commission

housing_catalyst_logo_236.png

Housing Catalyst in Fort Collins

APG-logo-with-ds-440x432-1.webp

APG Living & PMC Property Group

Information Architecture Evolution 

We transformed a sprawling sitemap into a lean, task-oriented structure by mapping user intents to core navigation pillars. The new architecture prioritizes immediate utility over institutional hierarchy.

Streamlined Categories

Consolidated 14 top-level menu items into 5 core navigation pillars based on user intent.

Clear Labels

Replaced jargon like "Section 8 Management" with human-centered language like "Rent Assistance."

User-Centric Hubs

Created specific entry points for Applicants, Residents, and Landlords.

Housing (1).png

Recommendations

Based on user feedback and testing (tree testing and card sorting), the following improvements were made to the site map to ensure better navigation and clarity:​

  • Streamlined Categories: Simplified grouping of content based on logical user flows.

  • Clear Labels: Ensured all categories and subcategories are labeled intuitively.

  • User-Centric Structure: Grouped content based on user personas and their specific needs.

Screenshot 2026-04-26 214025.jpg
  • Accessibility Enhancements: Integration of tools to support diverse needs.

  • Search Optimization: Implementation of features like autocomplete, query suggestions, and alternative search recognition.

  • User Assurance: Enhanced external login pages for trust and clarity.

Reflection

This project reinforced how much the framing of a persona shapes what you notice. Running the same website through Arthur's eyes and Jay's eyes produced completely different findings. Arthur's journey had friction but mostly resolved. Jay's never did. That contrast was the most compelling argument we could make for why IA changes, not just aesthetic ones, were what the site needed.


It also sharpened how I think about research artifacts as evidence rather than decoration. The tree test data and card sort results gave the recommendations weight they would not have had if we had just audited the site ourselves and offered opinions.

bottom of page