EV Charging
Integration

Problem

Too many apps, no shared pattern

In Azerbaijani fintech, every adjacent service still lived in its own app — its own login, its own registration, its own way of doing things. There was no shared UX pattern for something as ordinary as paying to charge an EV, so users were expected to relearn a new flow every time they needed one.

Garage's EV map already had traction — about 1,600 unique users open it monthly, with EV penetration in Garage running 3.4× the national average. But it was view-only: users could see stations and build a route, then had to leave for the iVolt app to actually pay and charge. 92% of users with EV intent never converted to a transaction.

The goal: turn that intent into a transacting flow inside m10 — station discovery, iVolt balance top-up, and session monitoring, without leaving the super-app or learning a new app's conventions. The charging payment market is roughly 1M AZN/month.

Garage section engagement data — internal analytics

That gap broke down into four concrete friction points:

• Intent without transaction. The EV map let users find stations, but every session still had to be started and paid for in the iVolt app — 92% of intent-stage users never converted.
• Balance & tariff confusion. Interviews showed users didn't understand iVolt balance, debt states, or per-kWh pricing — several expected billing by the minute, not kWh.
• No session visibility. Once charging started, there was no way to see kWh delivered, cost, or time remaining from inside m10.
• Disconnected accounts. A new iVolt account had to be created and linked separately, with no clear path to resolve a negative balance before charging.

Approach

How I got there

I sketched the first version of this flow solo, with no existing pattern in the local market to borrow from — MVP1 was in production about a month later. Before that build, the research below shaped what it needed to solve.

We benchmarked task flows across the charging apps already used in Baku — m10, Touch, iVolt, and Azerishiq — timing and counting clicks for the core actions: finding a station, starting a session, and paying.

Functionality comparison — clicks and time to complete core actions across m10, Touch, iVolt, and Azerishiq

View full research →

We then ran contextual interviews with EV drivers in Baku. A recurring theme: drivers didn't understand how iVolt balance, debt, or per-kWh pricing worked — several assumed the car itself handled billing, or confused kWh with charging minutes.

Interview synthesis — EV drivers' mental models around balance, tariffs, and charging limits

View full research →

MVP1 scoped to a single integrated partner — iVolt — with everything else routed to the provider’s own app. That scope, plus a blocked data dependency, shaped what shipped first.

We structured the flow in three phases: discovery (finding a station or scanning a QR code), balance (topping up iVolt through m10, including debt clearance), and session (starting, monitoring, and completing a charge). The balance step got the most iteration — it's the one place a user can get blocked entirely if the minimum isn't met.

User scenario — full journey from opening the map to completing a charging session

One early concept combined bonuses, iVolt balance, and the top-up amount on a single screen, toggling between the two balance sources right above the keypad.

Rejected concept — combining bonuses and iVolt balance on one top-up screen made it unclear where the money was coming from and where it was going

It tested poorly — people couldn't tell whether they were spending bonuses or iVolt balance, or which one the amount they typed would apply to. We separated iVolt balance into its own dedicated step, which is what shows up as the balance gate and top-up screens in the final flow below.

Final design

The complete flow

This was a big flow to design on its own, well before payment enters the picture — six steps covering the full EV charging journey, from map discovery through balance top-up to live session monitoring. Each is shown separately below.

Entry point — the m10 home screen and My Garage surface the iVolt balance and the Electric Charging Stations tile, leading into the station map and list.

Non-integrated providers — stations from networks like Gofar still show connectors and pricing, but starting a session hands off to the provider's own app.

iVolt — the first fully integrated partner. Users pay directly inside m10 and scan the charger's QR code to identify the exact connector.

Balance gate — a session can't start below 1 AZN. A negative balance (debt) is shown in red and blocks the Start button until it's cleared.

Payment confirmation — a successful top-up unlocks charging, followed by a confirmation once the session begins.

Live session monitoring — real-time kWh, cost, and duration, with the option to stop charging or check progress from My Garage.

Outcome

Results & status

MVP1 — the map, iVolt onboarding, balance top-up with debt clearance, and the full session lifecycle — is live in m10's Garage. MVP2, extending the same model to additional providers, is in progress and depends on resolving the blocked station-data contract.

iVolt is the first partner, not the only one. The goal for m10 is to become the single place EV owners handle charging end to end — station discovery, balance, and payment — across every provider in the market, so drivers stop needing separate apps for each charging network. MVP2 is the proof of that model with one partner before opening it up to the rest.