Recently, I took my car in to the repair shop. The car was running fine, as far as I could tell, so the only reason for the visit was because an indicator light had appeared on my dashboard – I got the check engine light. The light had been on for a long time, but I ignored it. Part of me felt the car was fine, that it was running smooth and it was something inconsequential, like time for a 70,000 mile checkup or something like that. Part of me, though, feared the worst. Something nefarious was going on under my hood, and the car could break down at any minute. The cost would be enormous. I was putting off service, because I didn’t want to address either of these possibilities.
And that’s the problem with the check engine light; it provides the minimum amount of information possible. As alarms go, it is a terrible alarm. All it tells me is that some problem exists in my car somewhere. The possibilities are seemingly endless. Because of this lack of information, now I have to have an “expert” diagnose the problem for me. Except for this expert is really just some diagnostic software that the car manufacturer created to decode the problem, which of course, costs me a not insignificant sum of money.
All told, this was a terrible User Experience (or is it Customer Experience???). Think for a second about how backwards this is. Something goes wrong in the car, which one in a vast array of sensors picks up. This sensor is triggered, which alerts the check engine light to fire. This indicator provides a tiny amount of information to me, the car owner. I then take time out of my day and spend my money to get the problem diagnosed. This may have made sense in the 1950s, when powerful computers were expensive, and not embedded into every car on the road. Back then, a physical sensor tripped and the best that was possible was a light on the dashboard (and drivers were thankful for it).
So let's look at the situation. A sensor identified a problem. Rather than dumb this down, the proper information could have been directly relayed to me. If the manufacturer didn’t want to waste space on the limited dashboard space, they could have sent the exact problem (in everyday language) to the console computer, which could have alerted me to the new problem. Then, I could have decided right then and there what course of action to take. I needn’t spend 6 months – yes it took me 6 months to get it checked out – worrying about what the problem was. The car manufacturer didn’t need to spend hundreds of thousands, if not millions, of dollars developing the software to decode the messages. And I didn’t need to spend my money to have them run this software.
The thing is, this type of alerting already occurs in my car. If tire pressure is too low, I get a message telling me so (of course it doesn’t tell me which tire). If a headlight is out, which just occurred, I get a message. Why does this not apply to every sensor in the car? What is the difference between tire pressure and broken sensor (which is what the problem turned out to be)? Is it because the manufacturer doesn’t want me to work on this area of a car? Do they not trust me to take the proper action? Do they just want to make some extra money off of me as a consumer who knows little about cars?
To make matters worse, it turns out that by the time I brought my car in, there were two problems. Two sensors stopped working. I had no idea. A second problem with the check engine light is that it is a group alarm. It encapsulates so many problems, yet it provides no indication of whether it represents one problem, two problems, or one thousand problems. It doesn’t tell me if I have a serious problem that I need to pull over right away for, or a minor issue that I can ignore. I can imagine my engine being on fire, and still all I see is the single check engine light.
After the diagnosis, I decided not to replace the sensors. Problem solved, right? Nope. I rolled out of the service area and the check engine light is still on. So even though I have chosen to ignore the message, it still stares at me every time I am in the car. And, if a new problem develops, I will never know about it, because the light can’t do any more that it is doing already.
This is the opposite of a good User Experience. I have not read or seen much about Apple CarPlay or Alphabet’s (still weird to not call it Google) Android Auto, or any of the newer in car consoles. My fear is that they are more focused on the infotainment then they are with helping me make better decisions with my car. I understand that problems don’t happen very often in a car, but when you are talking about decisions that are worth hundreds or thousands of dollars to the driver, a car that can make repair decisions clear will be worth its cost. No more wasted time, money, and worry trying to address a problem that may or may not be bad.
Now, let’s think towards the future and autonomous cars. These are going to introduce a plethora of new sensors and technologies. What happens when something goes wrong with one of these components? Will the car manufacturer replicate the check engine light, with a ‘check automation’ light? How safe will you feel in one of these cars not knowing what part of the automation is not working? I know how I would feel. It’s not good enough for observability into the automation, and it isn’t good enough for the rest of the car either.
We need to demand better out of our car manufacturers. Rather than adding in new features that demo great but that no one will use – which applies to most of what’s in the console – car manufacturers would better serve their customers that have been staring us in the face for decades.
And that’s the problem with the check engine light; it provides the minimum amount of information possible. As alarms go, it is a terrible alarm. All it tells me is that some problem exists in my car somewhere. The possibilities are seemingly endless. Because of this lack of information, now I have to have an “expert” diagnose the problem for me. Except for this expert is really just some diagnostic software that the car manufacturer created to decode the problem, which of course, costs me a not insignificant sum of money.
All told, this was a terrible User Experience (or is it Customer Experience???). Think for a second about how backwards this is. Something goes wrong in the car, which one in a vast array of sensors picks up. This sensor is triggered, which alerts the check engine light to fire. This indicator provides a tiny amount of information to me, the car owner. I then take time out of my day and spend my money to get the problem diagnosed. This may have made sense in the 1950s, when powerful computers were expensive, and not embedded into every car on the road. Back then, a physical sensor tripped and the best that was possible was a light on the dashboard (and drivers were thankful for it).
So let's look at the situation. A sensor identified a problem. Rather than dumb this down, the proper information could have been directly relayed to me. If the manufacturer didn’t want to waste space on the limited dashboard space, they could have sent the exact problem (in everyday language) to the console computer, which could have alerted me to the new problem. Then, I could have decided right then and there what course of action to take. I needn’t spend 6 months – yes it took me 6 months to get it checked out – worrying about what the problem was. The car manufacturer didn’t need to spend hundreds of thousands, if not millions, of dollars developing the software to decode the messages. And I didn’t need to spend my money to have them run this software.
The thing is, this type of alerting already occurs in my car. If tire pressure is too low, I get a message telling me so (of course it doesn’t tell me which tire). If a headlight is out, which just occurred, I get a message. Why does this not apply to every sensor in the car? What is the difference between tire pressure and broken sensor (which is what the problem turned out to be)? Is it because the manufacturer doesn’t want me to work on this area of a car? Do they not trust me to take the proper action? Do they just want to make some extra money off of me as a consumer who knows little about cars?
To make matters worse, it turns out that by the time I brought my car in, there were two problems. Two sensors stopped working. I had no idea. A second problem with the check engine light is that it is a group alarm. It encapsulates so many problems, yet it provides no indication of whether it represents one problem, two problems, or one thousand problems. It doesn’t tell me if I have a serious problem that I need to pull over right away for, or a minor issue that I can ignore. I can imagine my engine being on fire, and still all I see is the single check engine light.
After the diagnosis, I decided not to replace the sensors. Problem solved, right? Nope. I rolled out of the service area and the check engine light is still on. So even though I have chosen to ignore the message, it still stares at me every time I am in the car. And, if a new problem develops, I will never know about it, because the light can’t do any more that it is doing already.
This is the opposite of a good User Experience. I have not read or seen much about Apple CarPlay or Alphabet’s (still weird to not call it Google) Android Auto, or any of the newer in car consoles. My fear is that they are more focused on the infotainment then they are with helping me make better decisions with my car. I understand that problems don’t happen very often in a car, but when you are talking about decisions that are worth hundreds or thousands of dollars to the driver, a car that can make repair decisions clear will be worth its cost. No more wasted time, money, and worry trying to address a problem that may or may not be bad.
Now, let’s think towards the future and autonomous cars. These are going to introduce a plethora of new sensors and technologies. What happens when something goes wrong with one of these components? Will the car manufacturer replicate the check engine light, with a ‘check automation’ light? How safe will you feel in one of these cars not knowing what part of the automation is not working? I know how I would feel. It’s not good enough for observability into the automation, and it isn’t good enough for the rest of the car either.
We need to demand better out of our car manufacturers. Rather than adding in new features that demo great but that no one will use – which applies to most of what’s in the console – car manufacturers would better serve their customers that have been staring us in the face for decades.