Overview

Fast Match was Lyft's dedicated airport pickup experience — riders show a 4-digit code to a driver in a designated queue. The feature had a slightly elevated cancel rate driven by perceived wait time. I researched how to reduce uncertainty and cancellation without surfacing raw ETAs that discourage riders.

Research Setup

• My Role: Principal researcher for the Airports & Venues team
• Methods: Contextual interviews · On-site observation · Hypothesis-driven research
• Locations: LaGuardia · Chicago Midway · Portland

Core Insight

• Riders care about their own wait, not the system's — front-of-line riders want to know when the next car arrives; back-of-line riders need information to decide whether to stay or leave.
• A long ETA number was more discouraging than a long visible line: "Seeing 15 minutes is more worrying to me than seeing a long line."
• Riders not yet in the queue needed wayfinding, not timing information.

How Might We

"Give users enough information to calculate their own wait time as accurately as possible — without discouraging them from staying?"

Impact

• iOS FM cancel rate: −16.5%
• Android FM cancel rate: −10%
• Research readout drove PM, design, and engineering to actively participate in sessions, accelerating experiment design

Recommendations

Strategic ETA display · Contextual transparency for long waits · Comparative value vs. alternatives · Driver proximity metrics