Interactive Kiosks & Map Redesign — High Museum Wayfinding
Help visitors move confidently between buildings and exhibits, without asking for directions.
Team Muse Collective partnered with the High Museum of Art (Atlanta) to reduce wayfinding friction across its multi-building campus. We redesigned the on-site kiosks and map to clarify where I am, how to get there, and the best route for me.
Timeline
Aug 2024 - May 2025
My Role
Lead UX Design, User Research, Design System
The Team
1 Researcher, 2 Designers
OVERVIEW
Background
• Challenge
Visitors have struggled with navigation at the High for 20+ years, leading to confusion and missed exhibits.
• Opportunity
Understand why navigation fails and design a system that improves orientation, routing, and confidence.
Process
Final Design
• Way-finding Kiosks
These kiosks would be stationed across the museum to show where I am, provide routes to destinations, and highlight exhibits.
• Web App
Portable access to the same system without downloads; visitors can plan routes, filter by preferences, and access maps anywhere.
• Design System
Consistent iconography, building colors, signage language, and map layers for alignment across staff and visitors.
Impact
• Direction questions to staff
0%
0%
↓
Before: 6.8 questions/hour
After: 3.9 questions/hour
• Time to reach exhibits
0%
0%
↓
Before: Avg. 11 min
After: Avg. 8 min
• Visitor confidence navigating
0%
0%
↑
Before: 46% rated “confident.”
After: 62% rated “confident.”
• Navigation without staff help
0%
0%
↑
Before: 41% completed
After: 53% completed
PROBLEM
Framing the Problem Space
After preliminary research, we understood the problem more in-depth. 3 separate buildings, connected by only 3 bridges across limited floors, and inconsistent wall heights that distort spatial perception.
What makes it difficult?
And… visitors can't see these with current tools.
Problem Statement
How might we improve users’ experience with Navigating between buildings at the High Museum of Art?
RESEARCH
Breaking Down Research Questions
What is currently known about wayfinding & engagement?
Literature Review
What we did
• 40+ papers and articles
Key takeaways
• Visitor explorer motivations are common; wayfinding shapes engagement time.
• Movement design (paths, thresholds, decision points) strongly influences exploration.
• Navigation tools vary; each has trade-offs → kiosk + map must work together.
How do other museums guide visitors?
Comparative Analysis
Sample
• 10+ comparable multi-building museums
• 10 comparison categories
Key takeaways
• Mobile apps dominate digital guidance.
• Tours and paths are often preference-based (time, interests, accessibility).
• Physical signage tends to follow collection types.
How do visitors navigate at the High?
Field Observation
Protocol
• 4 days; 3h/day
Key takeaways
• Visitors are poorly supported by physical navigation tools.
• Visitors struggle to orient on maps; “Where am I?” is unclear.
• Many ask security for directions → the current system isn’t self-serve.
How do current tools shape the experience?
Contextual Inquiry
Stakeholder Interview
Participants
• 16 CI participants (20–70 yrs) · 2 interviewees
• 11 trips · 22 hours
Key takeaways
• Inconsistent signage and map language.
• Visitors can’t locate themselves; unaware of best routes for their preferences.
• Staff frequently reiterate directions; guidance varies by familiarity.
Translating Findings to Requirements
• Research Findings
• Design Requirements
Visitors’ spatial perception differs from the staff’s. |
DR1
Align mental models with unified spatial representation.
|
DR2
Support varying motivations (quick, deep, specific).
|
DR3
Reduce reliance on staff and support effective communication.
|
DR4
Provide routes and location indicators.
|
DR5
Establish a consistent vocabulary and visual system.
DESIGN
Ideation
• Co-design Workshop
We ran a co-design workshop to collaboratively explore navigation solutions with diverse participants. The aim was to leverage lived experiences, eliminate our own design bias, generate creative ideas, and converge on a platform that balances feasibility and user needs.
• Design Decision
After discussion with our client, we decided to move forward with interactive kiosks and web app as the primary solution. This direction balanced user desirability, stakeholder feasibility, and scalability, while still leaving room for layering in future physical or digital enhancements.
User Flow + Information Architecture
To structure the experience, we first mapped the end-to-end user flow for both kiosk and mobile.
The information architecture was designed to address three core tasks:
• Preset Routes
Provided the 3 most common routes and optimized routes for each one.
• Explore Page
Browse by type, building, or personal preferences.
• HeartMatch
Combined the existing "Museum Tinder" feature into our system to help with better decision-making.
Wireframing
• Low-fidelity Wireframes
We translated the flow into low-fidelity wireframes for both kiosk and mobile. These wireframes allowed us to test core navigation logic early without over-investing in visuals.
* These are only 1/5 of the wireframes
• 3D Map Exploration
To address DR1: aligning visitor and staff mental models, we first experimented with a 3D map representation of the museum. The goal was to make the complex, multi-building structure more immediately understandable than a flat 2D plan.
• Before
• After
• Resting Screen Animation
To help visitors quickly grasp the museum’s navigation system, we designed a looping resting screen animation for the kiosks. The animation reinforces the shared mental model between staff and visitors, while also inviting users to begin planning their visit with a clear call-to-action.
Design System
To create a consistent, scalable foundation for both kiosk and mobile, we built a design system that defined the visual language, interaction standards, and reusable components. This also helped the alignment across platforms and addressed DR5 (consistent vocabulary and visuals).
High-fidelity Prototype - Version 1
• User Scenarios
After validating the wireframes, we created our first high-fidelity prototype to test not only the flows but also the visual design system in action. To make the prototype realistic and focused, we designed around two representative user scenarios.
First-time user to Kiosk
A new visitor approaches the kiosk, completes onboarding, explores nearby collections, plans a cross-building route, and seamlessly continues the journey on mobile via QR code.
User Flow: Onboarding - Nearby collections - Route planning - Mobile
Design Goal: Ensure clarity for new users by making orientation and cross-platform continuity effortless.
Returning user to Kiosk
A returning visitor logs in, views personalized nearby collections, refines their route planning, and transfers updates to mobile for convenience.
User Flow: Login - Nearby collections - Route planning - Mobile
Design Goal: Reduce friction for repeat visitors and make personalization more valuable over time.
• Prototype Map
* The prototype contains 100+ screens across kiosk and mobile, covering both onboarding and repeat user journeys.
ITERATIONS
Testing & Evaluation Methods
To validate early navigation flows and ensure clarity
Wireframe Testing
Participants
• 10 visitors representing different demographics
Process
• Conducted wireframe testing with museum visitors using key navigation tasks (e.g., locating an exhibit, planning a route).
Outcome
• Identified early pain points such as unclear labels and the need for faster entry methods, which guided our first round of changes.
To identify usability issues through expert review
Heuristic Evaluation
Participants
• 10 experts (MS-HCI students and UX experts from industry)
Process
• Ten 45-minute sessions involved walkthroughs of kiosk/mobile flows with structured feedback and commentary
Outcome
• Led to 20+ improvement suggestions, including major ones like clarifying terminology, adding quick-start directions, and providing preset routes.
To assess visitor task performance on site
Task-Based Usability Testing
Participants
• 10 participants (a mixture of contextual inquiry participants and randomly sampled visitors).
Process
• Each 30-minute session included: participants performing navigation tasks across kiosk and mobile flows, followed by NASA-TLX and SUS questionnaires.
Outcome
• Users found the system supportive overall, though mental load was rated high.
• Led to 10+ improvement suggestions.
Iterations
Across three rounds of iteration, we made over 40+ changes, each improving clarity, accessibility, and navigation flow. Below are the major changes that shaped our final design.
• Quick Start Directions
Experts felt it was a cognitive load for users to recall how to get to their destination based on the map’s visual clue alone, hence, we added quick start navigation tips that provide a quick step-by-step guide to their destination.
Had to rely on visuals
Users relied only on visual map cues to remember directions.
Heavy cognitive load
Visitors felt unsure how to get to exhibits without asking staff.
Step-by-step support
Quick Start directions gave 2-3 clear and direct navigation steps.
Less need for staff help
Visitors felt more independent navigating without extra help.
• Floor (2D) + 3D Views
Experts and users both felt that having only a 3D view of the building would not assist users with getting them between specific locations. To help with that, we implemented floor views to provide more details for users as they navigate within the museum.
3D view only
Good for overall layout, but lacked clarity for floor-level details.
Ambiguity in routes
Visitors weren’t sure which exact floor or room an exhibit was on.
Improved precision
Users could see both macro structure and micro-level exhibit locations.
Dual views
Helps with the overall mental model of the museum.
• Preset Routes
Visitors with limited time felt overwhelmed having to plan routes from scratch, a pain point emphasized in both expert feedback and user testing. Families and tourists especially wanted a quicker option. We introduced preset routes, like “All Collections” or “Special Exhibits,” which are the two most common , to let them start exploring immediately without a long setup.
No shortcuts
Every visitor had to manually select exhibits to create routes.
Time-consuming
Quick visits (tourists, families) found this overwhelming.
Preset routes
Added shortcuts like “All Collections” or “Special Exhibits.”
Time-efficient
Visitors could start exploring immediately without heavy planning.
• Routes via Access Points
Users mentioned it was difficult to orient themselves when starting from less familiar areas of the museum. They wanted reference points tied to recognizable landmarks like elevators, restrooms, or cafés. To address this, we added “routes via access points,” allowing visitors to use well-known locations as anchors to confidently reach their desired exhibits.
No anchor points
Routes started only from current location on the map.
Low confidence
Visitors felt lost when the system’s starting point didn’t match their mental reference points.
Anchor-based routing
Added option to select familiar access points (e.g., elevators, restrooms, cafés) as route origins.
Increased confidence
Visitors could now plan routes using recognizable landmarks, reducing disorientation.
FINAL PROTOTYPE
Ready for Launch UI
After three rounds of iteration, we developed the final high-fidelity prototype that integrates all major changes and feedback. This version demonstrates how visitors can seamlessly plan, navigate, and explore the museum using both kiosk and mobile interfaces.
• Kiosk Screens
* These are only 1/10 of the final prototype
• Mobile Screens
* These are only 1/10 of the final prototype
REFLECTIONS
What I learned
• Balancing user needs with feasibility
One of my biggest learnings from this project was how to balance what users asked for with what was technically and visually feasible. Users often wanted more detail, more clarity, or more personalization, and I had to think critically about which changes were meaningful to implement immediately and which could be staged for later iterations. This helped me practice prioritization and taught me that design is as much about trade-offs as it is about creativity.
• Iterating based on evidence, not assumptions
Conducting wireframe testing, heuristic evaluation, and usability studies made me realize how often my first assumptions were wrong. Simple changes like adding legends, clarifying “Access Code” over “Username,” or providing quick-start tips had a much bigger impact than I anticipated. This reinforced the value of testing early and often—small usability improvements can transform the experience.
• Designing with real people in mind
What grounded this project was seeing actual museum visitors interact with the prototypes. Observing moments of hesitation or delight helped me empathize with users beyond abstract personas. It reminded me that design isn’t about perfect screens but about supporting real people in real contexts.
My Amazing Team
I'm very grateful to work with an incredible team of classmates, mentors, and our industry clients. Here are the very people who made wireframes, deadlines, and all-nighters actually fun!
Powered by 4-AM vending machine coffee, overflowing Figma layers, and never too many pixel-perfect tweaks.