Rebuilding an EV charging management platform for scale
EV Connect (part of Schneider Electric) is one of the largest EV charging management platforms in North America. As EV adoption accelerated, the platform needed to scale, and the legacy system it was built on couldn't take it there. I joined as the first product designer to help make that shift: rethinking the information architecture, establishing a design system the developers could build against, and working out how new functionality could keep being added as the product grew.
Once the development needs for the new platform were clear, the design system became the priority. Time was the constraint, so the goal was something efficient and practical rather than complete. I built it from scratch and grew it component by component as the platform demanded them.
This isn't a consumer product where people learn by using it every day. Sessions are infrequent and consequential, so every component had to explain itself. That meant more text than a design system usually carries, and components designed to communicate clearly rather than compress information into the smallest possible space.
The platform is white-labeled, which shaped how the design system had to work. It carries other companies' brands, so its own personality had to stay quiet: strong enough to hold the product together, neutral enough to adapt to whoever is using it. That constraint informed how the system was built, with components designed to stay recognizable across different brand treatments and the visual identity kept in a layer that could change without touching the structure underneath.
The people using this platform are professionals: site hosts, operators, network managers, each with different needs and a real job to do. We talked with them directly to understand how they actually work.
These are power users, and they needed the full range of control. But they're also controlling physical hardware, where a wrong decision has consequences in the world. The design had to hold both: simple enough to understand at a glance, powerful enough not to get in the way, and layered so complexity was available without being imposed. A large part of the work was helping people avoid mistakes in the first place.
We had a lot of data: usage patterns, faults, network efficiency, driver behavior. And we had direct conversations with site hosts, operators, network managers, and drivers. The value came from putting the two together, using the numbers to find where the problems were and the conversations to understand why.
That combination drove the decisions, and it mattered most on the driver side, where the volume of input was highest and the patterns were hardest to see without both halves.
The insights only mattered if they pointed somewhere. We tied the design work to a small set of business goals: reduce hardware downtime, reduce the need for customer support, and increase first-time charge success. Having those goals defined changed how we prioritized. Instead of improving the product generally, we used the data and the conversations to identify the specific areas holding those numbers back, and focused the design work there.
Over the years my work covered most of the platform.
Bulk actions and fleet management
Energy management
Pricing and pricing policies
Customer support
Station diagnostics
Subscriptions
Commissioning and pre-commissioning of stations
Partner APIs
Dashboards
Each came with its own users, its own constraints, and its own way of failing, which is part of why the design system mattered so much.
The platform serves professionals, but the whole experience has a driver on the other side. The networks depend on them, so their experience had to be part of the work too.
To understand the driver experience, I put myself in it. I drove to different stations in cities around the US, in different EVs and different conditions, across different scenarios, to see how the network actually behaves when you're depending on it: the frustrations, the small moments of doubt, what people actually value when a charge works, and the opportunities hiding in the gaps.
Embedding myself in the driver experience, charging on a roaming network in Portland, Oregon
Charging should be as simple as plug in, charge, leave. In practice, the technology and infrastructure aren't there yet, so the real design problem is what happens when it doesn't work.
Every piece of hardware behaves differently, and drivers arrive at a station with no way of knowing what they're about to deal with. Roaming adds another layer, since drivers also use stations from other networks where we have even less visibility. The focus became helping drivers understand what's actually happening when a station doesn't work, and making sure site hosts know about problems on their network rather than finding out from a complaint.
First-time success isn't only a design problem. A lot of it is technical, with causes spread across the hardware, the connectivity and the networks themselves. But on the driver experience side, the clearest opportunity was assistance in the moment: helping drivers work out what's happening, find their own way through it, and stay away from the actions that cause a session to fail in the first place.
The driver's goal is simple: charge and get on with the day. Everything I learned across those years pointed at the same thing, so I redesigned the app around that one outcome rather than the features around it.
The challenge is that the driver's experience runs across hardware and networks we don't control. Designing for that meant I had to work with incomplete information and find ways to make the experience feel dependable anyway.
I joined as the first product designer, working directly with multiple product teams and developers across several time zones. I developed the design system and led the redesign of the platform. Toward the end, I helped hire the company's first Head of Design and a small design team was created, while my focus shifted to the driver experience.
Leslie Mathews and Kayla Wilson