CarMax — Vehicle Preview Enhancements

Improving wholesale vehicle review workflows so dealers could inspect inventory faster, navigate images more efficiently, and make more confident purchase decisions.

Role Senior Product Designer
Team 1 PM, 3 Engineers
Timeline 6 months
Focus Workflow optimization, interaction design, usability research, UI refinement
Outcome +16% engagement, +25% profit margin per auctioned vehicle

Overview

CarMax Auctions is CarMax’s wholesale auction business, where licensed dealers browse and evaluate vehicle inventory before bidding. Vehicle preview pages play a critical role in that decision-making process, especially for dealers assessing exterior condition and damage from photo sets.

This project focused on improving the vehicle preview experience by reducing friction in how dealers navigated, enlarged, and inspected vehicle images. The goal was to make the experience faster, more intuitive, and more supportive of confident purchasing decisions.

Why this mattered

For dealers using the platform, vehicle previewing is not a passive activity. It is a repeated, high-frequency workflow tied directly to confidence, efficiency, and buying behavior. Small interaction issues inside that experience compound quickly when users are reviewing multiple vehicles in a single session.

Improving the preview flow had the potential to strengthen both user experience and business performance by helping dealers inspect inventory more effectively and move through listings with less friction.

My role

I led UX across discovery, research synthesis, and solution direction for the vehicle preview experience. I partnered closely with product and engineering to identify the highest-friction parts of the flow, validate opportunities, and design improvements that balanced usability gains with implementation constraints.

I was responsible for:

  • framing the problem space

  • conducting and synthesizing research

  • identifying key usability issues

  • shaping solution direction

  • refining the UI and interaction model with engineering

The problem

Dealers relied heavily on vehicle images to evaluate inventory quality, especially when reviewing exterior damage. But the image browsing experience introduced friction at critical moments in the workflow.

Users often had to:

  • repeatedly click through image galleries

  • open enlarged views to inspect damage

  • close those views and resume browsing

  • repeat that cycle multiple times per vehicle

That behavior slowed evaluation and added unnecessary effort to a task users performed over and over again.

Constraints

This work had to fit within several practical constraints:

  • delivery needed to happen within a roughly 6-month window

  • existing technical infrastructure limited how much of the gallery experience could be rebuilt

  • changes needed to improve usability without disrupting familiar dealer behavior too aggressively

  • solution scope had to focus on the most valuable friction points instead of attempting a full redesign


Research and insights

To understand where the experience was breaking down, I conducted usability testing, interviewed dealers, and reviewed user behavior patterns in the preview flow.

The research surfaced a few clear signals:

  • users clicked rapidly through image galleries, in some cases 15 to 20 consecutive clicks, suggesting friction in navigation

  • dealers frequently enlarged photos to inspect exterior damage, then had to close that view and continue browsing, which created a repetitive and interruption-heavy flow

  • time on page per vehicle increased from roughly 3 minutes to 5 minutes for users previewing cars, indicating more effort spent reviewing each vehicle

  • both power users and non-power users showed increased time in preview, suggesting this was not isolated to one user segment

These patterns indicated that more time spent was not necessarily a sign of a better experience. In this case, it often reflected a workflow that required too much effort to complete basic inspection tasks.

OLD FLOW

NEW FLOW

Key opportunities

Based on the research, I prioritized two opportunity areas.

1. Reduce image navigation friction

Users needed a smoother way to move through photos and inspect details without constantly breaking their flow.

2. Improve damage review efficiency

Because damage assessment was central to inventory evaluation, the preview experience needed to better support quick visual inspection and easier access to relevant images.

Design direction

I focused on improvements that could meaningfully reduce friction without requiring a complete rebuild of the preview system.

The design direction centered on:

  • making image inspection more fluid

  • reducing repetitive click patterns

  • improving continuity while browsing image sets

  • supporting faster review of damage-related photos

  • keeping the experience familiar enough to avoid unnecessary relearning

This was an important tradeoff. A more ambitious redesign could have restructured the full gallery model, but given the platform constraints and timeline, I prioritized changes that would create the most value in the most frequently used moments.

Final solution

The final experience improved how dealers navigated, enlarged, and reviewed vehicle photos. Instead of forcing repeated open-close cycles and interruption-heavy behavior, the updated flow supported a smoother path through visual inspection.

The solution aimed to:

  • reduce interaction overhead during image review

  • help users inspect damage more efficiently

  • improve momentum when moving through a vehicle’s photo set

  • create a more usable and lower-friction preview experience overall

In addition to improving the photo experience, the final design also supported clearer continuity in the browsing flow so users could stay focused on evaluating the vehicle rather than managing the interface.

Outcomes

The updated experience contributed to measurable business and product impact:

  • +16% engagement in dealer tooling

  • +25% improvement in profit margins per auctioned vehicle

  • improved support for both power users and non-power users during vehicle inspection

  • stronger usability in a workflow directly tied to dealer confidence and purchase decision-making

Reflection

This project reinforced how small interaction improvements can create outsized impact when they affect a high-frequency, business-critical workflow. When users are evaluating inventory at scale, every unnecessary click becomes expensive.

It also highlighted the value of pairing behavioral signals with qualitative research. The strongest opportunities did not come from aesthetics alone. They came from understanding where user effort was accumulating, then designing targeted changes that made the workflow meaningfully easier.

Next
Next

Google — Cases Admin Draft Annotations