Speculative redesign — educational purposes only. This is an unsolicited concept redesign of Moovit. It is not affiliated with, endorsed by, or produced in partnership with Moovit or Intel Corporation. All screenshots are from the public App Store listing and are the property of their respective owners. This case study is presented solely as a portfolio demonstration of UX research and accessibility design methodology.

Case Study · 02 · 2026 · Accessibility / Transit

AccessPath

Redesigning Moovit for low-vision and motor-impaired transit users

An unsolicited accessibility-first redesign of the Moovit transit app — built on participatory research with users with disabilities and a WCAG-grounded audit of every interaction in the current app.

Speculative Redesign
MoovitExisting app · 1.5B+ users globally (conceptual redesign)
My Role
UX ResearcherAccessibility specialist
Duration
12 Weeks2026
Platform
iOS · AndroidSpeculative concept
Screen Reader Support
Motor Impairment Design
WCAG 2.1 AA Compliance
Participatory Research
Gen AI Synthesis
About the Project

Accessibility claims don't always match the lived experience.

Moovit reaches over 1.5 billion users globally and claims WCAG 2.1 AA compliance on web and VoiceOver/TalkBack support on mobile. But participatory research with disabled users revealed a gap between stated compliance and lived experience — 100% of screen reader participants failed the core navigation task on first attempt, and 2 of 3 motor-impaired participants abandoned mid-task.

AccessPath is a participatory-research-driven redesign of the Moovit experience — grounded in real co-research sessions with users with disabilities, a WCAG 2.1 AA audit, and an accessibility-first rethink of every core interaction.

Role
UX Researcher · Accessibility designer · Solo
Methods
Participatory Research · WCAG Audit · Co-design
Users
Low vision (4) · Motor-impaired (3) · Both (2)
Type
Speculative redesign of a live global app
The Core Problem

Disabled transit users were designing their own workarounds.

8 of 9 research participants had never been consulted by a tech team. They had instead developed personal workarounds — alternative apps, sighted helpers, avoided routes — that compensated for the gaps in Moovit's accessibility.

Moovit has made accessibility investments — but participatory research found that the gap between WCAG compliance and real-world usability for disabled transit users remains significant. The redesign builds on Moovit's existing accessibility foundation and addresses the specific failure points surfaced by research with disabled users.

Research Goals

What this research set out to answer.

01
100%
What actually breaks?
Every screen reader participant failed the core navigation task on first attempt — despite Moovit's stated VoiceOver/TalkBack support. The research traced exactly why: gaps between WCAG compliance claims and real-world usability, at both interaction and information-architecture level.
02
8 of 9
Who hasn't been heard?
8 of 9 participants had never been involved in a tech research session. Participatory research wasn't just a method choice — it was a corrective act. The research had to surface priorities that no audit alone would have found.
03
WCAG
What's the minimum that moves the needle?
Rather than a complete redesign, define the fewest changes — landmark audio, target zones, persistent filters — that would pass WCAG 2.1 AA and resolve the observed task failures within the existing app's information architecture.
Research Methods

Participatory. Grounded. Audited.

👥
Participatory Research Sessions
9 participants across three groups: 4 low-vision users, 3 motor-impaired users, 2 with both conditions. Co-research in real transit context — stations, buses, real navigation tasks.
🔍
WCAG 2.1 AA Audit
Full audit of the Moovit app against all four POUR principles. 23 violations identified across the three core flows — route search, results, and guidance. Violations organised by severity and effort to fix.
🗺️
Task-Based Usability Testing
All participants given three identical navigation tasks with screen recorders and think-aloud protocol. Failure points mapped against WCAG violations to identify the highest-impact fixes.
Landscape Analysis (Gen AI)
Disability prevalence data and AT usage patterns across 12 cities synthesized using Gen AI in 3 hours — established the scale of the underserved population and existing accessibility standards across comparable apps.
📊
Cross-Group Pattern Analysis
Observational notes coded across disability groups — revealed "unclear ETA language" as a top-3 barrier across all groups, not just screen reader users. This drove the landmark audio specification.
Co-Design Workshops
Participants were shown wireframes and asked to mark what would and wouldn't work before any explanation was given. Several interactions were discarded entirely based on first-pass co-design feedback.
Design Process

Accessibility-first, not accessibility-added.

01
Audit
WCAG analysis
02
Participate
Co-research sessions
03
Synthesise
Pattern mapping
04
Co-design
Wireframes with users
05
Prototype
WCAG-compliant hi-fi
06
Validate
Re-test with participants
"

I don't need a perfect app. I need to know when to step off the bus without panicking.

Participant 04 · Low vision · Regular bus commuter

Key Findings

The failures were structural, not surface-level.

100%
Screen reader task failure
Every screen reader user failed the core route search task on first attempt. The app's custom map component had no ARIA landmarks — VoiceOver and TalkBack users couldn't orient themselves without visual context.
2 of 3
Motor-impaired users abandoned mid-task
Two of three motor-impaired participants gave up mid-task and called a friend instead. The core navigation task required 7 sequential taps on targets below Apple's 44×44pt HIG recommendation — significantly harder for users with reduced motor control.
8 / 9
Never consulted by a tech team
8 of 9 participants had never been involved in any kind of technology research or design session. Participatory research found entire barrier categories — anxiety about missing stops, unclear audio cues — that didn't appear in any existing accessibility audit.
Wireframes

Mid-fidelity — three core screens.

Wireframes
Mid-fidelity · three core screens
10:42
Plan a trip
Accessible routes only
Current location
Central Library
Today
Leave now
Recent destinations
Riverside Medical Center
Oakwood Community Center
Plan
Saved
Alerts
Settings
Route search
1
Persistent a11y toggle — survives sessions
2
44px+ touch targets for motor users
3
Recent destinations reduce typing burden
10:43
Routes to Central Library
Showing accessible routes first
3 routes found
Route 12 → WalkAccessible
18 min Ramp Lift
Route 7 → Transfer
24 min Stairs
Route 3 → Walk
31 min Step-free
Plan
Saved
Alerts
Settings
Route results
1
Accessible route ranked first — not buried
2
Ramp/lift info inline on every result card
3
Stair warnings distinguished without color yet
10:51
You're on your way
Next stop
Market Street Platform B
2 min · Ramp on left side
Your route
Board bus at Main St stop A
Exit at Market St, use ramp
Walk 3 min to Central Library
Audio guidance onTap to adjust
Plan
Saved
Alerts
Settings
Active trip guidance
1
Spatial landmark in ETA — "ramp on left side" not just "2 min"
2
Done / current / upcoming step states visible
3
Audio guidance persists — tap to adjust, not re-enable
Fidelity Progression

Mid-fi → Hi-fi — what changed and why.

Mid-fi → Hi-fi
Mid-fidelity
10:43
Routes to Central Library
Showing accessible routes first
3 routes found
Route 12 → WalkAccessible
18 min Ramp Lift
Route 7 → Transfer
24 min Stairs
Route 3 → Walk
31 min Step-free
Plan
Saved
Alerts
Settings
High-fidelity
10:43
Route results
Routes to Central Library
Accessible routes only
3 routes found
Route 7 → Transfer ⚠ 1 step
24 min ⚠ Stairs
Route 3 → Walk ✓ Step-free
31 min ✓ Fully step-free
Plan
Saved
Alerts
Settings
What changed and why
Mid-fi decisions
Greyscale — structure proven without color influence
Nav-bar title — position established
Darker border on accessible card — distinction without color
Pure white canvas — neutral baseline
Hi-fi additions
Navy blue system — header, active card border, button, nav all unified
Lora serif title — companion tone, not transit utility
Amber warnings, blue confirmations — urgency differentiated without alarming
#F5F7FA cool off-white — neutral base that keeps the navy system clean
WCAG gap report draft
Violations organized by POUR principle and severity — reduced reporting effort ~60%, freeing the accessibility consultant for recommendations
WCAG gap report draft
Violations organized by POUR principle and severity — reduced reporting effort ~60%, freeing the accessibility consultant for recommendations
Design Decisions

Three interventions. Each grounded in a failure.

Contextual Audio
Screen Reader Support
Landmark audio with spatial orientation
  • VoiceOver/TalkBack output includes spatial context — not just stop name and time
  • "Walk 200m north, then board at Bus Stop 14B — Grafton Street" instead of just "ETA 4 min"
  • ARIA landmarks added to map, route list, and guidance panel
  • Audio trip summary available on route selection — pre-journey orientation mode
Motor Accessibility
Touch Interaction
Large-target zones with motor shortcuts
  • All interactive elements minimum 44×44px — meeting Apple HIG guidelines and WCAG 2.5.5 Level AAA enhanced target size
  • Motor shortcuts: swipe-up to confirm, swipe-down to cancel — reduces sequential tap count
  • Sticky "I'm off the bus" button always visible during guidance — no hunting, plain language
  • One-tap accessibility filter on home screen — no buried settings menu
Route Filtering
Search & Results
Persistent accessible route toggle
  • Accessibility filter persists across sessions — set once, applies everywhere
  • Ramp and lift availability shown on every result card — not buried in route detail
  • Accessibility badge on featured route (blue pill, step-free confirmed) — immediately scannable
  • Warning chips for inaccessible segments inline — visible before committing to a route
Hi-fi Redesign
iOS · Concept
Accessible route results — redesigned
9:31
Route results
Grafton St → St Stephen's Green
Accessible routes only
3 routes found
Bus 46A ✓ Step-free
8 min ↑ Ramp ⬆ Lift
Luas Green ⚠ 1 step
6 min
Route 15 ✓ Step-free
14 min
Map
Routes
Saved
Settings
9:38
Next stop
St Stephen's Green
4 min · 2 stops away
Boarded Bus 46A at Grafton St
Exit here — ramp on left
Walk 120m to destination
Map
Routes
Saved
Settings
Working Prototype

Tap through the accessible flow.

Interactive prototype — tap through the flow
10:42
Plan a trip
Accessible routes only
Current location
Central Library
Today
Leave now
Recent destinations
Central Library
Riverside Medical Center
Plan
Saved
Alerts
Settings
10:43
Route results
Routes to Central Library
Accessible routes only
3 routes found
Route 12 → Walk ♿ Best
18 min ↑ Ramp ⬆ Lift ✓ Step-free
Route 7 → Transfer ⚠ 1 step
24 min ⚠ Stairs
Route 3 → Walk ✓ Step-free
31 min ✓ Step-free
Plan
Saved
Alerts
Settings
10:51
You're on your way
Route 12 · 4 stops remaining
Next stop
Market Street Platform B
2 min · Ramp on left side
Your route
Board bus at Main St stop A — done
Exit at Market St — ramp on left
Walk 3 min to Central Library
Audio guidance onTap to adjust
Plan
Saved
Alerts
Settings
11:09
You've arrived
Central Library
18 min · Fully accessible · No stairs
Trip summary
Route Route 12 → Walk
Duration 18 min
Barriers None ✓
Plan
Saved
Alerts
Settings
Screen 1 of 4 — Search
Flow steps
1Search + toggle
2Pick a route
3Active guidance
4Arrive
Research link
Spatial landmark language ("ramp on left side") came from 100% failure rate on screen reader task — "two stops" gave no orientation.
Tap routes or the search button to navigate · Toggle the accessibility switch on screen 1
Evidence Base

How each result was measured.

Unlike CareSync, AccessPath's outcomes are not projections — they are direct measurements from prototype re-testing with the same participants who tested the existing app. Here's what was observed and how each figure was recorded.

100% → 0%
Screen reader task failure rate
1

Baseline from existing app testing: All 4 low-vision participants and both participants with combined conditions (6 total) failed the core route search task on first attempt with the existing Moovit app. Failure was defined as inability to complete the task without researcher assistance. Screen recordings confirmed zero successful unaided completions.

2

Re-test with the AccessPath prototype: The same participants were given the identical task with the AccessPath hi-fi prototype. All 6 completed without assistance. Root cause was the missing ARIA landmark structure — the prototype's ARIA roles and landmark audio gave screen reader users the orientation cues the existing app entirely lacked.

3

Think-aloud validation: Participants narrated their navigation in both sessions. In existing app tests, every participant verbalised confusion at the custom map component. In the prototype re-test, none did — confirmed in coded think-aloud transcripts.

3 taps
Maximum taps to complete route selection
1

Tap-count audit of the existing app: The motor-impaired usability session included a tap-count log alongside screen recordings. The existing route search → results → select flow required a minimum of 7 sequential taps. Several targets were below 44×44px — Apple's HIG recommended minimum — making them particularly difficult for users with reduced motor control. WCAG 2.5.8 (AA, WCAG 2.2) sets a 24×24px floor; WCAG 2.5.5 (AAA) recommends 44×44px.

2

Prototype design constraint: The AccessPath interaction model was designed to the 3-tap target — swipe gesture shortcuts, persistent accessibility filter, and consolidated route cards all reduce sequential decisions. The 3-tap figure is the maximum path counted across all task scenarios in the hi-fi prototype, not an average.

3

Validated in co-design workshops: All 3 motor-impaired participants completed the route selection task in the prototype without abandoning — compared to 2 of 3 who abandoned entirely in the existing app session. One participant completed with a single swipe gesture.

WCAG AA
All 23 audit violations resolved
1

The audit baseline: A full WCAG 2.1 AA audit of Moovit's three core flows — search, results, and active guidance — identified 23 distinct violations across all four POUR principles. The AI-assisted report (which reduced documentation effort by ~60%) organised them by severity and effort-to-fix to prioritise the redesign scope.

2

Violation-by-violation resolution: Each of the 23 violations was mapped to a specific design decision in the prototype — whether ARIA landmark structure, colour contrast ratios, tap target sizing, or ETA language clarity. Resolution was verified criterion-by-criterion during the handoff review, not claimed as a blanket pass.

3

Scope note: Compliance was verified at the prototype design level. A full engineering implementation would require additional automated and manual testing (e.g. axe-core, screen reader regression) before a live WCAG AA claim could be made.

8 / 9
Would switch to AccessPath
1

Exit interview question: At the end of each hi-fi prototype session, participants were asked a single direct question: "If this were available as an app today, would you switch from your current transit app to use it?" 8 of 9 said yes without qualification. 1 said they would use both depending on journey type.

2

Context of the 9th response: The one participant who gave a qualified answer used Moovit specifically for its social trip-sharing feature — a feature not in scope for this redesign. Their hesitation was unrelated to accessibility functionality, and they rated the accessibility improvements as "much better" than the existing app.

3

Unprompted qualitative feedback: Multiple participants made unsolicited comments during testing — "this is what it should have been from the start," "I've never been able to do this without help before." These were not prompted responses and were coded separately from the exit interview question.

These results come from prototype testing with 9 participants — a small, purposive sample. They demonstrate that the design decisions resolved the observed failures for this cohort. Large-scale validation, automated accessibility testing on a live build, and longitudinal retention data would be required before these results could be generalised.

Prototype Testing Results

Measured in testing. Grounded in evidence.

100% → 0%
Screen reader task failure rate — from 100% failure on the existing Moovit app to 0% in re-testing with the same participants on the AccessPath prototype. Landmark audio and ARIA restructure resolved every observed failure directly.
3 taps
Maximum taps to complete route selection in the prototype — down from 7+ in the existing app. Meets Apple's HIG 44×44pt target size recommendation and was validated across all 3 motor-impaired participants in testing.
WCAG AA
All 23 violations identified in the initial audit were resolved in the prototype design — verified against WCAG 2.1 AA criteria as part of the design handoff review.
8 / 9
Participants said they would switch to AccessPath from their current transit app if it were available — recorded in post-session exit interviews across all 9 research participants.
Reflection

What participatory research found that audits couldn't.

Audits find what breaks. Research finds what matters.

The WCAG audit found 23 violations. Participatory research found the anxiety of missing a stop, the shame of blocking the aisle, the workaround of calling a sighted friend. Neither alone would have been sufficient. Together, they produced a redesign that passed both tests.

The Gen AI WCAG gap report.

Violations organized by POUR principle and severity using AI-assisted reporting reduced the audit documentation effort by ~60%, and let the time saved go into more research sessions. The AI wrote the report. The researcher decided what to fix first.

Up next

A Date with the Owls — ASU Burrowing Owl Conservation

View case study