Turning broken dashboard into a product fleet managers can actually use : Linkbycar
LinkByCar is a French fleet intelligence platform. It gives operators — insurance companies, EV rental fleets, logistics companies, driving services — real-time visibility into their vehicles: location, battery health, driver behaviour, charging patterns, predictive maintenance alerts.
The platform serves anyone running a fleet. A company managing 30 rental EVs. An insurer pricing policies based on actual driving data. A logistics operator monitoring 200 trucks across three countries.
When I joined, none of this was functional.
The problem
The company had an early product and a vision. What it didn't have was a usable interface.
A previous designer had attempted a component library — partially built, nothing interactive, nothing that a user could actually operate. The dashboard, the centrepiece of the product, was a full-screen map. No hierarchy, no action prompts, no path forward. Users landed there and stalled.
Underneath the visual problems was a more fundamental one: nobody had defined what a fleet manager actually needs to do when they open the app. The product had data. It had no opinion about what mattered.
There was no information architecture. No design system. No process for introducing new features without breaking existing ones.
My Approach
Making "close your eyes" work on a screen
I didn't start with screens. I started with a question: what does a fleet manager need to accomplish in the first 60 seconds of opening this product?
The answer was almost always the same — navigate to a specific vehicle and complete a task. Check its battery. Review a safety alert. Pull a trip log. Everything else is secondary.
That one insight restructured the product. If the fastest path to a vehicle is the core job, then everything on the dashboard should either serve that path or get out of the way.
Structure
Rebuilt the sidebar from scratch — what features live together, what belongs at the top level vs. nested, what a fleet manager needs once a day vs. once a quarter. This was the foundation. Every screen decision followed from it.
Dashboard
Replaced the map-as-homepage with a surface that leads with status: how many vehicles are active, which ones have alerts, what needs attention now. The map became one view among many, not the default landing state.
Feature design
Once the core was solid, I introduced new feature lines. Battery Care Score and Vehicle Safety Scoring, both of which required translating dense telemetry data into something non-technical operators could read and act on.
Dashboard Redesign
The redesigned dashboard surfaces what matters: vehicle status at a glance, active alerts, quick access to any vehicle. The map is still there - it's just no longer the whole story.
Battery Care Score
EV fleet operators need to know if their vehicles are being charged correctly, whether battery health is degrading, and when a replacement is coming. Previously that meant reading raw data. I designed a scoring system - built on State of Health metrics, charging habit analysis, and degradation trends - that reduces all of that to a score, a trend line, and a recommendation.
Vehicle and driver Safety Scoring
For insurance clients and rental operators, driver behaviour is a liability question. I designed a safety scoring system that processes speed, harsh braking, cornering, and acceleration into a per-driver risk score - actionable for fleet managers, integrable for insurers.
Design System
50 components, built from scratch. Buttons, tables, alert states, map overlays, scoring cards, data visualisation primitives. Before this existed, every new feature was a coin flip. After, new screens could be assembled consistently in a fraction of the time.
Marketing website
I also designed and built the current linkbycar.com in Framer — product positioning across Fleet Management, Insurance Claims, and EV Solutions, with clear paths to demo requests and API documentation.
Results
Specific usage data is under NDA. Fleet manager productivity - how quickly operators can find a vehicle, assess its status, and act - has increased 2-3x since the redesign.
The engagement started as a project. It became a long-term retainer because the product got measurably better and kept moving.
"At GoMecano, we are revolutionizing car maintenance by bringing the garage to your home, but we were lacking knowledge about vehicles before intervention. I was blown away by the technology that Linkbycar deploys, which allows us to accelerate GoMecano's growth thanks to data."
Alexandre Nivesse
GoMecano
What this engagement actually was
Most designer-client relationships have a clear boundary: here's the brief, here's the delivery. This one didn't work that way.
I scoped features. I pushed back on bad ideas. I decided what to build next. When the founders needed to communicate the product to investors or enterprise clients, I shaped how it was presented. The retainer held not because I executed well — though I did — but because I made the product better than it would have been if someone was just following instructions.
That's the difference between a designer and a fractional product lead. If you're an early-stage team that needs one person to own both, this engagement is the most direct evidence of what that looks like.
