Self-identification: telling the driver what you need before the car arrives

Riders and drivers with visual and hearing disabilities had to explain themselves at every pickup. I designed self-identification for both sides of the trip, live globally after privacy review under GDPR. Refused trips fell sharply.

Live

Global

Rider screens for hearing and vision self-identification, each next to the driver’s view while waiting for the rider.

Project details

Role: Senior Product Designer — Product Inclusion (design lead for the initiative)

Team: PMs, engineers, UX research, content design, data science, legal, privacy, operations and support

Timeline: Announced in October 2024 with a worldwide rollout

Impact: Refused trips fell sharply, as did driver deactivations for discrimination

Scope: Self-identification for riders and drivers with visual and hearing disabilities, plus pickup preferences

・・・・・・・・・・・

The problem

Riders who are blind, have low vision, are deaf or hard of hearing often had to explain themselves by message or call at every pickup, the most failure-prone moment of the trip. Drivers arrived without context, and some accepted a trip only to cancel once they learned more.

・・・・・・・・・・・

What I designed

Rider self-identification: riders choose to tell Uber about a visual or hearing disability, and stay in control of what is shared.
Driver self-identification: drivers can say they are deaf or hard of hearing, so riders know how to communicate.
Guidance for the other side at the right moment, before arrival and not at the curb.
Pickup preferences that turn free-form requests into clear instructions for drivers, later reused in Senior Accounts.

Five rider screens for preferences and requests, including pickup preferences such as calling out the rider’s name and offering help to get in and out of the vehicle.

・・・・・・・・・・・

Privacy as part of the design

Disability is sensitive personal data. The feature went through privacy and legal review under GDPR before it reached Europe, and disclosure is always the rider’s choice.

・・・・・・・・・・・

Results

Refused trips fell sharply after launch.
Driver deactivations for discrimination fell.
Shorter wait for a driver who completes the pickup: fewer drivers accept and then cancel.
Live globally, in every language.

I tracked two signals: refused trips, and the ratings these riders received from drivers. A falling rating means drivers need better guidance. Results are described by direction only; the figures are internal.

・・・・・・・・・・・

Key learnings

Structured context reduces friction more effectively than reactive messaging.
Design for dignity, not for exception handling.

・・・・・・・・・・・

🔒 Additional details

Detailed flows and internal metrics are not shown in this public version. A deeper walkthrough is available upon request.