Toll Management for Fleets
I cross-referenced 3 data sources to find where fleet managers got stuck and redesigned every flow of the product.

The problem
The data showed errors and dead clicks in the flows, drop-off points and features with no adoption. Toll billing was sent by email. Inside the product, users saw statements and billing forecasts and could buy more tags, entering billing data and quantity.
The decisions
- I cross-referenced data from Google Analytics, Microsoft Clarity and Userpilot to find errors, dead clicks, drop-off points and features with no adoption.
- I redesigned all product flows, including tag purchase, statements and billing forecast. I used usability heuristics, business goals and user goals as the basis.
- I ran co-creation sessions with PMs and POs to prioritize features and improvements.
- I prototyped in Figma and tested the new flows remotely with one client (6 participants). I validated the decisions with the VP and product management.
What I did not do
I did not interview fleet managers before redesigning. The decisions came from data, heuristics and business and user goals. The test came afterward, with the prototype.
How I used AI in this project
I used ChatGPT to sketch ideas, support the analysis of results and insights and, in part, for wireframing.
Screens and artifacts


Current state
The prototype got positive qualitative feedback in the test. The project did not move to development and stayed at prototype.
What I would do differently
Escalate the lack of prioritization to leadership earlier, instead of insisting alone on moving forward with the teams.