Jan - May 2025 · Internship
Designed new on-the-lot mobile features to help 45M+ monthly buyers navigate hidden fees and dealer negotiations.
CarGurus is the #1 most visited automotive shopping site in the U.S. However, the company’s mobile app was rarely used at dealerships so myself along with three other Tufts human factors engineering students set out to design new mobile features. We wireframed multiple ideas and I created our final finance calculator prototype using Lovable.
The finance calculator was shipped to 45M+ monthly users!!
83% of user testers would recommend the feature to a friend
4.3/5 average ease of use score from on final prototype
On the business side, less than 10% of users opened the CarGurus app while at dealerships. As for buyers, 80% reported being stressed out by hidden fees and unclear pricing while only 9.7% felt confident they got the best deal. Existing calculators failed to estimate the true ‘out-the-door’ cost.
Based on CarGurus internal research and conversations with recent buyers, we decided to focus on improving information transparency surrounding financing and hidden fees. We also had to ensure our solution didn’t jeapordize dealers’ relationships with CarGurus.
The final prototype gives buyers a clear breakdown of all costs, simplifies financing steps, and serves as a meaningful resource for negotiations. I led the design direction across five iterations and built the final mobile prototype with Lovable.
Final prototype under NDA. See what shipped.
Research
Internal research showed the app was seldom used at dealerships, while secondary research and 5 buyer interviews revealed struggles with hidden fees and unclear financing terms. These findings highlighted cost and information transparency as a gap to address.
– Car Buyer Interviewee
We visited 3 dealerships, interviewed 5 dealers, and observed buyers struggling with financing paperwork. Buyer's order sheets scattered key information across multiple boxes, leaving buyers unsure if they were getting a fair deal.
After consolidating 9 buyer interviews, 5 dealer interviews, and multiple on-the-lot observations, we mapped the entire car buying process. Buyers often lost confidence at each step, eroding trust in dealers and leading to stressful decision-making.
Prototypes
Our team brainstormed and sketched 9 potential solutions, from feature comparison tools to AR visualization. This exploration allowed us to consider a full range of potential directions before committing to one solution.
By informally testing with buyers and pitching to CarGurus, we discovered which concepts resonated with users. We mapped our findings onto an impact vs effort analysis and selected 3 concepts to build out in Figma.
Generate, scan QR codes
Estimate out-the-door price
Vehicle price range by area
Accurate vehicle details
Rank and compare features
Book dealer visits upfront
Explanations for key terms
View cars side-by-side
Visualize cars and add-ons
I built low-fi interactive flows in Figma and tested them with 6 participants. Feedback showed the Finance Calculator gave buyers the greatest clarity and confidence, so we chose it as the key feature to prototype.
Deal-O-Meter
Finance Calculator
Estimate Summary
Design
I led the design for the Finance Calculator, frequently testing with participants and refining the visual design to address user needs. Through feedback, I added features like a sticky finance banner, a monthly payment goal, and options to save or share estimates.
Our Figma prototype conveyed the core flow, but complex interactions and live calculations weren't possible. At CarGurus' suggestion, I built a prototype in Lovable and later refined it in Cursor. Bringing the design to life with AI tools was an amazing moment that made me really excited about the future of building software.