A cabin interface now carries tasks that used to be mechanical. Decide task by task which ones the screen may own, and what happens when it cannot.
India’s electric vehicle market reached about six billion dollars in 2024 on a reported 36% compound growth rate, and buyers now ask about driver assistance and interface quality alongside range. Those figures set the commercial context; they do not settle a single design question inside the cabin.
The controls moved, the responsibility did not
Turning a key, shifting a gear and reading a fuel gauge were physical, single-purpose and available without power. Their replacements are software surfaces that mediate several tasks at once and depend on a boot sequence.
This is an authority question before it is an aesthetic one. Name, for each task, whether the driver, the vehicle or the software decides, and how the driver is told which one is in charge at that moment.
Do not migrate a control to a screen because it fits there. Migrate it when the screen version is faster to operate while moving, and safe when it fails.
Classify every cabin task before designing it
Four categories are enough to structure the decision, and the fourth is where most programmes are thin.
Route, media, climate presets: operable in one glance or by voice.
Drive modes and regeneration: the screen sets, a physical control confirms.
Anything needed under load, in gloves, or with the display dark.
What the driver can still do when the software is unavailable.
Assign every task to exactly one column, then test the fourth.
Four categories of cabin control: screen owned, shared, physical, and degraded operation.Test the degraded column on the road, not in a spreadsheet. A car whose cabin becomes unusable during an update is a product failure regardless of how the interface reviews.
Charging is part of the interface
Range anxiety is mostly an information problem. The vehicle knows its state of charge and its consumption; the driver needs the consequence of both, in the units of the journey they are actually on.
Show arrival state of charge rather than remaining kilometres alone, name the conditions that would change it, and be explicit when a charger is unverified. Confidence comes from being told the truth early, including when the answer is unwelcome.
Treat the charging network as a dependency you do not control. Design for a charger being occupied, offline or slower than advertised, because that is the common case.
Assistance features need visible boundaries
Driver assistance sells the vehicle and is then operated by someone who did not read the manual. The interface has to make the system’s current authority and its limits legible without instruction.
Whether the system is active, in one unambiguous indicator.
What it needs to stay active: lane markings, speed, weather.
How control returns, with enough warning to take it.
What it will not do, stated before the driver relies on it.
If a driver has to infer any of these, the feature is not ready to ship.
Four disclosures required of a driver assistance interface: engagement, conditions, handover and limits.Market position in this segment currently follows accessibility as much as capability. Simplicity a first-time buyer can operate is a defensible strategy, not a lesser one.
Decide what to validate next
Pick one journey, instrument it end to end, and measure task time while moving, error recovery, and what drivers do when the screen is unavailable. Then decide which tasks move back to physical controls.
Interface quality is now part of the purchase decision. Assess it against defined driving tasks rather than inferring it from a showroom demonstration.
Written by the studio, from work in delivery
