Silver Aero: finding, booking and paying for EV charging in one app

Problem

The hard part of an EV road trip is the charging stop, not the drive

Every driver I spoke to enjoyed driving electric. The stress came at the charger. Maps show stations that turn out to be broken, busy or built for a different plug. Payment happens in yet another app, and there is no way to know in advance how long a stop will take.

Drivers who rent EVs have it worse, because they don't know the car's connector or range. I set out to make the charging stop as predictable as a fuel stop.

Research

Drivers trusted their cars but not the charging network

I ran a 14-question Google Forms survey for people who own or rent an EV and shared it on social media. It collected 83 responses in a week. From the suggestions people left, I recruited four drivers for semi-structured interviews (three online, one in person) in the UK, Canada and India, including a dealer who sells EVs.

I transcribed the interviews, broke them into notes on an affinity map and built empathy maps, two personas and a journey map. Three insights shaped the design.

Framing

How might we make a charging stop as predictable as a fuel stop?

My point-of-view statement: a driver who would rather spend time at the destination than in a queue needs to find the charging station with the shortest wait. That led to three How Might We questions: how might we find the nearest fast charger, help renters understand EVs, and encourage local electricians to provide home charging?

Decision 1

One app for find, book and pay, instead of a layer on top of Google Maps

Interviewees switched between Google Maps, Zap-Map and Apple Maps to plan a single stop. My competitor review of Google Maps, Tesla, Volkswagen and Zap-Map showed why: none combined a trip planner, charging alerts and payment. Of the four, only Zap-Map took payment.

One idea on the How Might We board was to build on the map apps people already use. I rejected it because the failures happened at the hand-offs between apps. Silver Aero puts the route, nearby compatible stations, live availability and booking in one flow. The trade-off is a bigger app that would need real network data to be trustworthy, which a prototype can't show.

Decision 2

Pay with Apple Pay and keep the ticket in Wallet

Payment came up again and again. Interviewees described failed payments in proprietary charging apps, and the journey map showed Brandon waiting to pay at the station. Instead of building another payment engine, I used Apple Pay, which drivers already trust, and let them book a connector and a time slot in advance.

The confirmation becomes an Apple Wallet ticket with a QR code, so drivers don't have to reopen the app at the charger. The trade-off: this ties the concept to iOS. An Android version would need Google Pay and Google Wallet.

Decision 3

Set up the car once, so drivers only see chargers that fit

In my survey, only 22.9% of 83 respondents could say for sure whether a given charging station could charge a Tesla Model X, and 28.9% didn't know any of the charging standards I listed. One sticky note on the affinity map simply said 'We need unification of charging standards'. Renters were the most exposed.

During onboarding the driver picks their car by make and model, or scans it. Silver Aero then filters stations by compatible connector and explains the three UK plug types in plain language. I chose a one-time setup over asking for the plug type on every search. It costs an extra onboarding step, and the 'scan using AI' option is speculative and would need technical validation.

Decision 4

Add a service for hiring a local electrician to fit a home charger

In my survey, 53% of 83 respondents preferred a standard AC charger at home to a public charging station (47%), and the EV dealer I interviewed often had to find electricians to fit chargers for his customers. Drivers who move house face the same problem. So I added a Services tab: browse nearby electricians with ratings and hourly rates, book an installation or repair, see an itemised estimate, and pay with Apple Pay. Breakdown recovery sits on the same tab.

This was the riskiest scope decision. A marketplace needs vetting, insurance and supply on both sides, and the prototype doesn't solve those. I kept it because it answered a need from my research that no competitor covered.

Solution

The finished prototype: plan, charge and pay without leaving the app, and continue in the car

I built the phone prototype in Figma (60+ screens) and the in-car view in Adobe XD, following Apple's CarPlay patterns. The car screen shows turn-by-turn directions, the next charging stop and battery range, so the driver doesn't need the phone. The visual language is dark with teal accents and SF Pro, so it reads well at night and in a dashboard. Brands in the screens (Tesla, Audi, Apple, RAC, Google Maps) are only illustrative. The project has no affiliation with them.

Validation

Moderated tests: every task completed, with fixes needed to sizing and type

I tested the hi-fi prototype in moderated Zoom sessions of about 20 minutes, after participants signed consent forms. My project report records six participants, each working through seven scenario tasks, and every task completed in the time allowed.

Asked about the look and feel, 80% said they were satisfied and 20% said they would prefer a light mode. These are the figures as recorded in my project report: 80% and 20% of six people aren't whole numbers, so I can't now say exactly how many people each figure covers. The main issues were component sizes, typography and touch targets, which I listed for the next iteration. This was a small prototype test, so it is directional, not statistical. It also couldn't test the hardest part: whether real availability data would be accurate enough to trust.

Reflection

What I'd do differently

I'd narrow the scope earlier. The electrician service and breakdown recovery weakened the core charging flow, and they deserved their own study. I'd also record time on task and a System Usability Scale score for every participant, and retest after fixing sizing and type.

A real version would need live data from charging networks, legal and payment partnerships, and an Android version. I'd also design for when things go wrong: what happens when a booked charger is out of order when you arrive.