Indoor navigation for university students.
GBDA 210 · UX research and design for a campus navigation app with AR and indoor wayfinding.
Timeline
8 Weeks
Role
Research
UX/UI
Visual Design
Prototyping
Tools
Figma
SUS questionnaire
Team
Joy Zhang
Amy Liu
Megan Pilkey
Sandy Srinivasan
Overview
The Challenge
Indoor navigation gap
Existing apps stop at the building entrance, leaving students lost inside complex buildings.
Research Methods
12 user interviews
Contextual inquiry + affinity diagram + cognitive walkthroughs with real students.
Design Approach
Crazy-8 → Lo-fi → Hi-fi
Iterative prototyping in Figma, validated through 6 usability tests.
Key Outcome
Indoor navigation that goes beyond GPS
Turn-by-turn indoor directions with AR overlay, schedule sync, and accessibility preferences.
PROBLEM
Students get lost the moment they walk through the door.
Once a student steps inside a building, no navigation tool guides them to the right room, floor, or hallway. At a campus like University of Waterloo, with tunnel systems, multi-building complexes, and hundreds of numbered rooms, that gap causes real daily stress.
The impact isn't just inconvenience. Students leave 20 minutes early "just in case." They visit buildings the day before to preview rooms. They plan routes the night before class. Navigation was quietly consuming time students couldn't afford to lose.
How might we help students navigate seamlessly from anywhere on campus to a specific room inside a specific building?
This raised a question
MY CONTRIBUTION
This was a 4-person team project. Here's how my contributions broke down.
What I owned independently
Proposed the typography, colour, and component system adopted across the 21-screen prototype
Designed the indoor navigation flow, including highlighted paths, turn-by-turn guidance, arrival state, and the map options screen
Ran an independent cognitive walkthrough, identifying a missed usability issue
Defined functional requirements and conducted a moderated usability test
Collaborated on
Co-created the interview guide and research plan
Empathy mapping and personas
Crazy-8 ideation and lo-fi sketching
Hi-fi prototype (21 screens split across 4 members)
Usability testing (1 session each)
Final presentation delivery
RESEARCH
Interview Design
Our team ran 12 semi-structured interviews — 3 each across 4 members. Before writing a single question, we made a deliberate choice about how to ask, because framing changes everything.
The guiding principle:
ask about what happened, not what they want.
Users are not always reliable predictors of what they want, but they are precise when describing past frustration. So instead of asking “what would help you navigate better?”, we asked “tell me about the last time you were late to class.” That shift shaped what we learned.
Interview Findings
Four themes emerged consistently across all 12 interviews and shaped every design decision that followed.
Affinity Diagram
HMW
Existing maps fail at the door.
Everyone used Google Maps or Apple Maps and everyone had a story about a wrong entrance, a longer route, or a hallway with no guidance. The tools work outdoors and stop there
Students need to see the route, not read it.
Participants independently described the same things: AR arrows, entrance photos, highlighted floor paths. The absence of visual cues was the problem, not text instructions.
Navigation is a hidden time tax.
Students were losing time they couldn't afford — leaving early just in case, previewing buildings the day before, planning routes the night before class.
Students navigate by landmark, not coordinate.
No one said "go north." Everyone said "turn right at the DC Library" or "past the vending machines." How students describe campus bears no resemblance to how navigation apps give directions.
"The map shows you where the building is — but not the classroom. I want it to say: walk 10 metres, turn left, it's on your right."
2nd-year UW student, interview participant
From Finding to Feature
What it does for the user
Finding
Design decision
Students needed visual, floor-level guidance
Students used two apps for schedule and navigation; left early due to timing uncertainty
Students think in landmarks, not cardinal directions
Some participants needed to avoid stairs — for others, a hard requirement
AR mode with turn-by-turn overlays, highlighted paths, and landmark anchors
WATIAM schedule sync on the home screen with ETA per class
All directions use landmark-anchored language throughout
Accessibility preferences set once, applied to every route automatically
Navigate an unfamiliar building without needing to know the layout first
Check your next class and navigate from one place — no app-switching
Directions match how students already think about campus
Accessible routing is always on — nothing to reconfigure
Research also drove what we cut.
Route Sharing
No participant habitually shared navigation routes. Adding it would have cluttered the UI for a behaviour that doesn't exist in practice.
Static Layout Map
Duplicated what Live GPS already offered, and most buildings already post their own floor maps. More cognitive load, no added value.
Social feed
Students were already time-pressured. A social layer would compete for attention at exactly the moments when focus matters most.
Ideate
With our research findings defined, we diverged before converging — exploring multiple directions before committing to a solution.
Crazy-8 Sketching
As a team, we each rapidly sketched 8 ideas in 8 minutes, focusing on improving the indoor navigation experience. This surface a wide range of directions — AR overlays, landmark cues, arrival states, and schedule integration — before we narrowed down.
Design Process
01 Discovery
12 interviews + empathy maps + secondary research on campus navigation
02
Define
Personas, user tasks, HMW statements, and a prioritized requirements list
03
Ideate
Crazy-8 sketching; explored GPS, AR, and layout-map navigation modes
04 Prototype
Lo-fi wireframes → high-fidelity Figma prototype across 21 screens
05
TEST
1 pilot + 6 usability tests with SUS scale and open-ended follow-ups
→
→
→
→
SOLUTION
A campus navigation app that bridges outdoor GPS and indoor wayfinding, designed specifically for the university environment.
01
Live GPS Navigation
A highlighted floor path updates in real time, with landmark-anchored turn-by-turn directions and estimated arrival time always visible.
Cognitive Walkthrough
After the team reviewed the prototype internally, I ran a cognitive walkthrough with a separate participant — because when you already know how something works, you stop seeing where it breaks.
Issue: No way to go back.
The participant stopped exploring the moment they couldn't find a back button. A missing back button doesn't just inconvenience users — it makes them afraid to interact at all.
Fix: Back arrow added to every screen, so students can always return to where they came from without losing their place.
USABILITY TESTING
Protocol:
1 pilot test, then 6 moderated sessions. Screen + voice recording. SUS questionnaire + open-ended follow-up questions.
3 tasks:
Check your schedule and read out your courses for the day
Find the preferences section in your profile and turn off stairs
Switch between Live GPS and AR navigation modes
What testing revealed:
Familiar interfaces build confidence.
Recognisable icons reduced hesitation — users navigated an unfamiliar app without friction because the patterns felt known.
Live updates felt trustworthy; static maps felt passive.
Users responded positively to the real-time route updates — the app felt responsive and reliable as they moved through the building.
Indoor navigation was the standout feature.
Every participant mentioned it unprompted when asked what was most useful — validating the core thesis of the project.
Schedule integration reduced cognitive load.
Users appreciated having route planning connected to their class schedule. The built-in notification prompts helped them plan departures without switching apps.
REFLECTION
What I'd do differently
Our participant pool skewed toward GBDA students, who are design-literate and likely more articulate about navigation problems than a typical user. I'd recruit more broadly across Science, Engineering, and graduate programs to surface different building contexts and habits.
I'd also run at least one in-context session, having a participant walk to an actual location while narrating their experience. Memory compresses and edits, but observation doesn't.
What this project reinforced
The most important skill in research isn't asking good questions. It's knowing what to do with the answers, which findings to act on, which to set aside, and which mean cutting something you worked hard to build.
