Uber Supplier Portal Dashboard

Background

In regions like Latin America, India, and Europe, many Uber drivers don’t own their vehicles. Instead, they rent them from third-party fleet operators. To manage these relationships, Uber provides a Supplier Portal that helps fleet owners track performance, manage drivers, and scale their operations.

However, a critical piece was missing: real-time visibility into business performance. Fleet owners were relying on manual reports from Uber’s operations teams, which were time-consuming to produce, inconsistent across markets, and difficult to act on. Our challenge was to create a dynamic performance dashboard within the Supplier Portal. The goal was to give fleet owners the insights they needed to coach drivers and grow their business.

The Challenge

Some of the key challenges included:

• The legacy constraints of the Supplier Portal UI, which was not built for dashboard-style layouts

• Misaligned stakeholder assumptions, such as “more data is always better,” that conflicted with UX best practices

• Ambiguity around how users would interact with performance data at scale

• A strong need for rapid validation to inform and align on a clear design direction

Design Process

I served as the Sr. Product Designer leading this initiative from start to finish. This included initial concepting, wireframes, interaction design, prototyping, and cross-functional alignment.

I began by exploring layouts in Figma, identifying what data needed to be surfaced and how it should be grouped. After identifying what content/elements/components needed to exist I created various explorations on layout of the interface. I worked closely with product managers to balance business goals with user needs. I also partnered with engineers to ensure feasibility and clarity.

I presented options to stakeholders (PM, Eng, greater design team) and there was debate over interface direction and assumption of how a "power user" might interact with the product. It was apparent that the static screens were not able to fully convey the complex interactivity of the interface and it became important to validate those interaction patterns early to help guide design decisions.

Prototyping

I decided to quickly prototype the interface (using Uber's internal React Front-End framework) to simulate the experience in a browser context in order to get more buy-in and demo different options to be critiqued. This reduced cycle time and allowed for feedback from users faster to continue to improve quickly and efficiently.

User Testing

I conducted remote usability tests with fleet owners, including one who managed over 300 vehicles and had built his own fleet management tools using Uber’s API.

Key insights included:

• Users preferred a dedicated performance hub over scattered charts

• Inline chart interactions were confusing and difficult to parse

• Fleet owners wanted to compare individual drivers to fleet averages and access key metrics like mileage and utilization in one place

• Owners preferred sending pre-written messages to drivers for coaching, rather than making phone calls

These insights directly shaped the dashboard’s hierarchy, layout, and features.

Refinement

Using the feedback, I restructured the experience into a multi-tiered system:

• A high-level dashboard with key fleet metrics such as earnings, trips, utilization, and driver quality

• Drill-down views for Trips, Drivers, and Vehicles, each offering targeted performance summaries

• A dedicated driver detail view to review performance, benchmark comparisons, and enable coaching actions

Prototyping enabled us to quickly test layout variations, validate flows, and refine interaction patterns. Once the design direction was confirmed, I returned to Figma to finalize and annotate the designs for handoff.

Final Design

Impact

This project led to a dashboard that was:

• Validated by users for clarity and usefulness

• Easily supported by engineering through well-defined and feasible design specs

• Aligned across product, design, and data teams

By validating patterns early, we avoided unnecessary scope expansion and focused development on building a solution that worked.

Reflection

This experience reinforced the importance of grounding UX work in actual user behavior, especially when working with complex data. Early testing and design flexibility helped us:

• Align more quickly as a team

• Avoid subjective debates about layout and interactions

• Deliver a tool that genuinely fit the workflows and mental models of our users

If I had more time, I would have conducted more iterative testing across regions and begun codifying the design patterns into reusable dashboard components.