Skip to main content
Uber logo

Schedule rides in advance

Reserve a rideReserve a ride

Schedule rides in advance

Reserve a rideReserve a ride
Data / ML, Engineering

Maximizing Process Performance with Maze, Uber’s Funnel Visualization Platform

August 16, 2018 / Global
Featured image for Maximizing Process Performance with Maze, Uber’s Funnel Visualization Platform
Figure 1. This simplistic view of a funnel highlights where user drop off might occur.
Figure 2. The Maze Sunburst visualizes how the tool tracks and organizes data from driver sign-up sessions.
Figure 3. In this visualization, we look for events whose names end with START, SUCCESS, or FAILURE. When we add this filter, the shape of the sunburst becomes very different.
Figure 4. Using sequence filters, we can view specific sign-up sequences, such as failure events that follow successful events.
Figure 5. A spire of purple nodes at the top of our visualization represents clicks on a browser’s back button for a certain web page involved in the driver sign-up process, enabling us to diagnose conversion traps.
Figure 6. Maze casts light on specific parts of the driver sign-up funnel.
Figure 7. The Maze app architecture incorporates a React client, a Node server, and a monitoring dashboard.
Figure 8. The Maze service architecture leverages several different data sources to aggregate and filter for a given query.
Figure 9. Arcs of the same angle occupy more space in the sunburst visualization as radius of the circle increases.
Markov chain viewSunpulse animation between two sunbursts
Variable width sunburstDifferent coloring modes
Individual sequences view, clustered by similarity
Figure 10: This set of experimental visualizations we developed in Maze shows data insights from a variety of different perspectives.
Yujia Luo

Yujia Luo

Yujia Luo is senior software engineer on Uber’s Growth Insights Platform team. He previously worked on Uber’s mobile feed system, focusing on optimizing its performance and developing its architecture.

Jerome Cukier

Jerome Cukier

Jerome Cukier is a senior software engineer on Uber’s Visualization team. He specializes in information design and building visual tools.

Posted by Yujia Luo, Jerome Cukier