I travel a bit for work, and when the trains don’t work out, I’m able to get hire cars – which is a great convenience. The experience of getting a different hire car each week over the past few months has revealed to me just how important the human computer interface is becoming, and just how poorly it is currently being engineered. And in a chunk of metal travelling at high speeds, a poorly engineered interface can be deadly. I know this.
Even though I’m 6’4″ I’m generally quite happy to take the economy option that pops up in the corporate booking system. I don’t get to choose the brand or model, so it’s always interesting to see what turns up.
Also I know I’m an edge case – not many people jump between brands and models like this – they usually develop brand loyalty and learn their car over a number of months/years. But I suspect this will change as cars become servitised and manufacturers begin delivering mobility services rather than car ownership. More on that later.
Anyway, this is usually how my user-experience goes:
- I sit down and adjust the seat and mirrors. These things are designed pretty standard these days, so no heavy cognitive load.
- Then I put my feet on the pedals, hands on the wheel and start the car. I get a feel for the engine in an instant, from the depth of whine/purr/grumble. I can quickly identify what stick does what – wipers, gears, indicators, lights. Generally that plastic and metal stuff is pretty standard as well
- Then – before I even get in gear – I have to face the new modern bit: I want to connect to the car. From my experience over recent months, this is where the car manufacturers are all struggling and where they are missing a trick.
- And then I have the challenge of operating this interface whilst actually driving. This is the bit that is really worrying. Screens destract rather than inform. My eyes are off the road too long. And the complexity of the system drives frustration that impairs overall concentration. Bluetooth connections fail, speech recognition fails to make calls, simple tasks like changing music stations involves confusing icons and interactions.
The variability of interfaces is enormous, and the number of quality designs is very very limited. Invariably there’s a crude touchscreen, and invariably the interface is an electronic version of the push button plastic of before. No effort to rethink the functions and features. And trust me, a touchscreen is far worst than a dial for scanning radio frequencies. It’s a case of new technology damaging old behaviour.
And often the data and interfaces are split into different locations – data about the car’s performance is set next to the dials in front of you, whilst all media related data is separated into the middle of the dash – no idea why. And then you get the controls set into the steering wheel. All vary in function and location. I mean – what the hell does this button do?
It’s my firm belief that these interfaces will become the most important factor influencing car purchases in the future. Engines are going to become more efficient and less about performance. People will expect them to start every day, handle well and have a low cost of ownership. Most cars perform very well on these terms. There was a narrow range in physical and cost performance between all the models I drove (the VW was better than the Vauxhall), but I suspect it is getting narrower as technology improves.
The interface is the new consumer battleground, and much needs to be done.
Here are some use cases we would all like to see realised. First the basics:
- I can actually make a call by calling out the person’s name. The number of times I’ve tried saying names at different volumes and dialects, only to get a computer lady telling me to repeat the name is mind bending. Interface Rage.
- RDS, TA, Region – all these legacy radio acronyms disappear and the function they serve is made apparent and intuitive. I have no idea what they do anyway, but I know if I hit one of them annoying traffic people keep interrupting songs
- Just as pedals and dials are now consistent, the location of interfaces needs to becomes consistent – data is collected in one safe place, so it informs and doesn’t distract. Gestures are employed so you don’t have to look to find the thing to point at. I can adjust volume by moving my finger up and down a screen – left to right to change stations or tracks. No more swerving whilst stabbing fingers dangerously at touchscreens. Like this.
- iPhones no longer come with a lengthy and costly instruction manual. Cars do – often with a separate one for the interface. Who wants to read these things? Who does? Design them to be intuitive, save the cost of the paper.
Now onto the more interesting and advanced stuff that can help brands build loyalty. These are more important to me than anything that gets talked about on Top Gear. I’m sure I’m in an addressable segment, but can’t believe I’m alone in it. And with Gen Y recognised to be less about ownership, and more about reduced environmental impact, I’m confident we’ll see a growth of this segment. Any service that can take effort and hassle out of people’s lives will succeed. This is being proved time and time again. So if a brand can achieve the following things, I will buy a car from them.
- I’m running low on petrol, so the car directs me to the nearest petrol station
- It’s lunchtime so I’m likely to be hungry, so the car directs me to chains en route where I’ve previously eaten
- When I stop for petrol, remind me to top up the washer fluid
- The car auto-diagnoses problems and notifies me of what I should do. I feel like a caveman having to still check the oil dip stick. Tell me when it’s low.
Unfortunately the law of car retailers is that these are cool new things and therefore must go into the premium models first. I think that’s a mistake. Design a good interface across all models – though yes, make additional features modular for different buyer segments.
And Landrover are doing the same with their higher-end models, though sticking to car-themed chrome styling, which I don’t think is helping anyone.
And of course Apple is getting into the market with CarPlay which looks promising, though unclear who the partners are. And of course Google is in there with OAA that doubtless will mesh with its driverless car and robotics investments, which seem to be mired in health and safety regulations. Personally, I hope that the open source MirrorLink will prevail. Trusted brand VW are putting it into new Polos, which will help. Having just moved from an apple to android handset, I think the configurability of open source will work better across the variety of models and brands.
But lets go further than just the interface, with one more future scenario that I am personally very keen on:
In the near future I don’t own a car – instead I pay a subscription for a car service. Most of the time we as a family use an electric car for local journeys (school drops offs, visits to friends and relatives, shopping trips), upgrading to a hybrid MPV for a two week summer holiday in Devon. In this world the vehicle has been reduced to a component of a mobility service. I am more interested in the purpose it serves that the thing itself – i want to get around in different situations which require different vehicles. The vehicle itself is expected to perform but is ultimately temporary and disposable.
In this situation the interface becomes my connection through the product to that service. And the service knows my driving habits, how to position seats and mirrors and the radio stations I like. It has my contacts stored in the cloud so I can make calls easily. Upon key contact (or phone contact?) all of these things are set up in whatever vehicle the service provider has provided.
Whether Audi or Avis end up offering this service, I don’t know, but it’s coming. Peugueot are playing with Mu, though it doesn’t look heavily invested.
But let’s be honest – this example is not about interface design – it goes way beyond that into service design, the implications of which are an entirely new business model that disrupts entire supply chains. And given the challenges Tesla are facing in the US trying to bypass the car retailing consortia with their direct to market sales model, I don’t hold out hope for rapid change. However for me this is scenario is as inevitable as the rise of peer-to-peer file sharing and OTT TV services.
Customer demand is like water – it always finds the path of least resistance and over time will erode rock to get what it wants. It will be no different for cars.