Uber app on a phone laid over a vintage city map
Accessibility Research Qualitative Methods Co-design DePaul University

Improving the Rideshare Experience
of Users Who Are Blind or Have Low Vision

A qualitative research study exploring the barriers BLV individuals face when using Uber, and co-designing solutions to make rideshare more equitable for everyone.

4
BLV participants interviewed
83%
of guide dog users denied rideshare access
3
major challenge categories identified
4
co-designed solution features

Project Overview

The brief

My Role

UX Researcher & Co-author

Methods

In-depth Interviews, Qualitative Analysis, Co-design

Tools

Zoom, NotebookLM, FigJam

Focus Area

Accessibility, Inclusive Design, Transportation

Ridesharing apps like Uber and Lyft promise independence and mobility — but for users who are blind or have low vision (BLV), the reality is often far more complicated. Despite relying on these services more heavily than sighted users, BLV passengers face systemic barriers that existing research had not fully addressed with actionable solutions.

This study set out to change that. Through in-depth interviews with four completely blind Uber users, we surfaced the real friction points in their rideshare experience — and worked with participants directly to co-design solutions that could meaningfully improve their journeys.

Methodology

How we approached this

01

Literature Review

Surveyed existing research on BLV rideshare experiences, identifying gaps where actionable solutions were lacking — particularly around driver behaviour, pick-up logistics, and app accessibility.

02

Participant Recruitment

Recruited four participants who were completely blind, had used Uber within the past five months, and lived in the Chicagoland area or nearby. All participants were compensated for their time.

03

In-depth Interviews via Zoom

Conducted one-hour virtual interviews per participant, using a pre-written script for consistency. Two moderators per session — one leading, one observing and taking notes.

04

Qualitative Analysis with Loose Coding

Transcribed all sessions, then used NotebookLM for initial loose coding. Organised codes into thematic buckets on FigJam, followed by a second manual analysis to draw conclusions.

05

Co-design Sessions

Invited participants to brainstorm solutions to their biggest pain points collaboratively, ensuring the resulting recommendations reflected real user needs rather than assumptions.

Who We Spoke To

Our participants

Participant 01

72-year-old woman · Chicago, IL
Completely blind · Cane user
Primary app: Uber

Participant 02

39-year-old man · Houston, TX
Completely blind · Cane user
Primary app: Uber

Participant 03

61-year-old woman · Chicago, IL
Completely blind · Guide dog user
Primary app: Uber

Participant 04

74-year-old woman · River Forest, IL
Completely blind · Guide dog user
Primary app: Uber

All four participants used Uber as their primary rideshare service, often at a subsidised cost through government assistance programmes. Despite this cost benefit, each had significant accessibility challenges that affected their safety, confidence, and independence.

Data Analysis

Affinity mapping

After transcribing all four interviews, we used NotebookLM for loose coding to surface recurring themes, then organised codes into buckets on FigJam. The affinity map below shows how participant data was clustered — from demographics and sentiment to specific problem areas and co-designed solutions.

FigJam affinity map showing participant data clustered into problems, solutions, sentiments, and demographics

Research Findings

Three major challenge areas

Across all interviews, three interconnected challenge categories emerged. Each one eroded participants' sense of safety, trust, and independence in different but compounding ways.

Communication Breakdown

Drivers who didn't speak English, who communicated nothing at all, or who cancelled without reason left participants stranded — sometimes in dangerous situations. Six consecutive cancellations in the rain was one participant's reality.

Pick-Up & Drop-Off Failures

Drivers expected BLV passengers to visually locate them. They honked instead of speaking, asked riders to describe building colours, and frequently dropped passengers at incorrect locations — insisting it was the right place.

Driver Insensitivity & Discrimination

Participants were yelled at, sworn at, and discriminated against for having guide dogs. Over 83% of guide dog users nationally report being denied rideshare access — a crisis of compliance with accessibility law.

App Accessibility Gaps

Screen readers encountered unlabeled buttons that read only "button, button, button." GPS-based location cues were visually displayed but not conveyed to assistive technologies. The ride cancellation button was difficult to find.

"I've had drivers show up and not speak at all. They just pull up and don't speak — they assume that I see them. I want them to know what I need from them. I need them to speak."

Participant 03, Chicago

"There's some issues on the app where it'll just say button, button, button and it's not clear on what it is — so their app should be labeled better."

Participant 03, on Uber's screen reader compatibility

Co-designed Solutions

What participants designed

Rather than imposing solutions, we invited participants into the design process. The four features below emerged directly from collaborative brainstorming sessions with the people most affected.

01

Haptic Proximity Feedback

Phone vibrations that gradually intensify as the driver approaches — providing a spatial, non-visual signal that works even in loud environments where VoiceOver can't be heard. Particularly valuable at crowded pick-up points like hospitals or transit hubs.

02

En Route Audio Notifications

Spoken updates throughout the trip — announcing passing landmarks, turns, and major streets — so passengers feel informed and confident they're heading to the right place without relying on a driver who may not communicate.

03

Frequent Pre-Pickup Updates

More granular status updates while a driver is en route — including contextual messages like "stuck in traffic" or "delayed by train." Participants described wanting to know not just minutes away, but why — especially when the countdown stalls unexpectedly.

04

Mandatory Accessibility Training

Structured training for drivers on how to assist passengers with visual impairments — including verbal communication standards at pick-up, guide dog etiquette, and legal obligations. Recommended by three of four participants as the single highest-impact change.

Recommendations

What we recommend

01

Implement Automated Messaging & Communication Features

Build transparency into every stage of the ride — from en route updates and pre-pickup status messages to haptic arrival cues. Automation reduces dependence on individual driver behaviour, creating a more reliable baseline experience for BLV passengers.

02

Provide Additional Support, Services & Training for Drivers

Driver training must address disability discrimination, guide dog accommodation (a legal requirement), and standardised pick-up and drop-off procedures for BLV passengers. Include translation support for language barriers. The variability in driver behaviour is a systemic problem — and training is a systemic solution.

03

Commission a Formal Accessibility Report for the Uber App

Unlabeled buttons, poor screen reader compatibility, and a difficult-to-find cancellation flow are symptoms of an app that hasn't been rigorously tested with assistive technology users. A formal accessibility audit would identify and prioritise the most impactful UI changes — benefiting millions of users globally.

The curb-cut effect at work

Every solution we co-designed with BLV participants would also benefit sighted riders. Automated status updates reduce anxiety for everyone. Driver training creates more empathetic service across the board. Easier cancellation flows help all users. Accessibility isn't a niche concern — it's a design quality that lifts the whole experience.

Reflection

What I learned

This project deepened my understanding of what it means to design with people, not just for them. Co-design sessions with participants produced more specific, more feasible, and more nuanced solutions than anything we could have generated independently — because the people who live these challenges every day know exactly where the system is failing.

I also came away with a sharper appreciation for how structural inequity shows up in product design. The barriers BLV passengers face with Uber aren't edge cases — they're the result of design decisions that defaulted to sighted users as the norm. Making the invisible visible in this research felt like exactly the kind of work UX should be doing more of.

If I were to extend this study, I'd want to conduct a comparative analysis between Uber and Lyft — Lyft's accessibility-forward guidance for drivers may translate to measurably different passenger experiences, and that insight could be powerful advocacy material for both platforms.

Want to see more?

Explore other projects in my portfolio, or get in touch to talk research, collaboration, or anything in between.