What we got wrong about idle fees in build 2

The problem with showing a moving target

When we set out to show true total cost before arrival, idle fees seemed straightforward. Find the charger's rate, add it to the per-kWh charge and parking, multiply by your estimated session time, and show the driver the real number. No surprises at the charger.

In practice, idle fees are less a fixed cost and more a tax on uncertainty. A 10-minute rapid charge might incur zero idle penalties. Forty minutes at the same charger? Different story. And the thresholds vary wildly across the 40+ networks Volta aggregates. Some networks don't charge idle at all. Others begin after 45 minutes. Some scale it per minute after a threshold. We were collapsing all of that complexity into a single figure, and we were doing it based on an assumption about how long the driver would stay.

The Nottingham driver wasn't angry about the fee itself. She was unsettled because Volta had claimed to show her the real cost before arrival, and the real cost had shifted the moment she deviated from the plan. We'd solved for transparency but created a new kind of uncertainty.

Why we thought showing an estimate was enough

Honestly, we were fixing a different problem. Before build 2, drivers had no visibility into idle charges at all. They'd arrive at a charger, top up, and discover weeks later that their statement included a £3 or £5 idle fee they hadn't budgeted for. We wanted to end that surprise. The estimate felt like a huge step forward.

What we underestimated was how many drivers arrive at a charger with a flexible plan. A commuter might expect 15 minutes but stay for 30 because a call runs long. A delivery driver might finish a job faster than expected. A parent waiting for a child might kill time. In those moments, an idle-fee estimate becomes a moving goalpost, and instead of reducing uncertainty, we'd just shifted it to the charger itself.

The fix wasn't to estimate better. It was to be clearer about what we couldn't know.

After reading that message and a dozen like it, we rebuilt how idle fees appear in the app. Instead of showing a single cost based on an assumed session time, we now show the idle-fee structure for that specific charger upfront. If a network charges no idle fees, we say that clearly. If idle kicks in after 45 minutes, we show it. If the rate scales, we show the brackets. Journey planning still calculates a route-level estimate for your trip, but at the charger level, the app now gives you the rules rather than a prediction.

It sounds like a small shift, but it changes the relationship between driver and app. We're no longer claiming to know how long you'll stay. We're saying, 'Here's what this charger costs if you're quick, and here's what it costs if you linger.' The driver makes the call. The surprise is gone.

What this taught us about complexity

The UK's 40+ EV charging networks aren't a unified system. They're a patchwork of operators, each with pricing logic built for their own equipment and strategy. Volta's job is to unify that view without erasing the differences. That's harder than it sounds, especially when you're trying to give a driver a single number to trust before they arrive.

The idle-fee rebuild forced us to accept that some complexity can't be reduced without losing accuracy. A more honest product shows you what matters, even if it means asking you to make a judgment call. That's riskier from a design standpoint. It's easier to give people one number and let them blame the network if it's wrong. It's harder to say, 'This is the rule; you decide if it fits your plan.'

We're a young app. We're still learning where transparency ends and overwhelm begins. But build 2's idle-fee mistake taught us that the answer isn't to be more clever with our estimates. It's to be more honest about what we don't know.