Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
  • The latest TMC Podcast (#14) is now available on YouTube and all major podcast networks. We covered FSD Beta's exciting v11 update, Enhanced Autopilot coming to the U.S. and Canada, and more!

A Digital Designer's POV on V11

V11- Spotify scrubbing compact.gif


Long time lurker, first time posting here! I've had a Model 3 since May '18, and I've been a professional Designer (sometimes called UI/UX Designer and Product Designer) for several years now. I didn't expect Tesla to take so many basic steps back in driver usability, nor did I expect that as we approach April, Tesla hasn't reverted more changes.

From the POV of design, looking at usability, interaction, and accessible design, I wrote a case study detailing why V11 is a step back for basic usability in a Tesla.
(Btw, the prose and context there is for a very general audience and the design community vs terminology and context all of us here are highly familiar with)

TL;DR There’s three primary issues with V11’s design:​

  1. Burying numerous, previously glanceable and reachable items into menus such as “Quick” Controls requiring extra taps, and removing heated seats and defrost controls from the bottom bar
  2. Ergonomics, or lack thereof, for the new alerts and gestures that are blocked by the steering wheel
  3. Homogenized UI from previously distinguished buttons/actions, making for a greater cognitive load from the driver when needing to distinguish between a grid of buttons
I get the sense that there have been two different design leaders between V9 and V11. There's clear evidence of the lack of Information Architecture weighting as so many things like charging and tire pressure monitoring were buried in the Controls menu despite having differing importance in usage. That means Tesla's digital design team chose form over function with a lot of V11 choices.

Personally, the heated seat and media scrubbing are my two daily annoyances.

Semi-TL;DR:​

I have a Twitter thread with visual examples of the more prominent V11 design issues.

Here's to hoping Tesla hears us, and improves the design - and hopefully maturing their approach to digital design! It doesn't mean they still can't shake up in-car digital experiences as they have for 10 years. But to move forward now with more thoughtfulness on usability for a much, much larger and diverse ownership base.
 
Welcome to TMC, and thanks. I've been designing hardware that has to interact with software for IDK, 25 ish years, and I've basically come to all the same conclusions. Sadly, I've seen a massive degradation is management's understanding of what it takes to integrate good UX/UI design with hardware functionality, across all industries. Few companies make the space in the schedule anymore to really get it right. Good designers are quitting in the face of the chaos. I know this to be the case at Tesla, as I've known 4 engineers personally who quit, one worked there since the 2000's until last year, the other 2 for shorter periods. All 4 quit in the face of increasing stress and chaos.

Tesla's development strategy is in a class by itself. When anyone can spin off a new idea, basically self test on the side, and roll into production, with apparently zero regression testing, all kinds of weird stuff can get rolled out.

I don't have a lot of hope for anything. Heck, in a recent update, somehow a nice new feature, the headlight card pop up on stalk push, got killed. Who knows why, maybe an accident, maybe on purpose?
 
Welcome to TMC, and thanks. I've been designing hardware that has to interact with software for IDK, 25 ish years, and I've basically come to all the same conclusions. Sadly, I've seen a massive degradation is management's understanding of what it takes to integrate good UX/UI design with hardware functionality, across all industries. Few companies make the space in the schedule anymore to really get it right. Good designers are quitting in the face of the chaos. I know this to be the case at Tesla, as I've known 4 engineers personally who quit, one worked there since the 2000's until last year, the other 2 for shorter periods. All 4 quit in the face of increasing stress and chaos.

Tesla's development strategy is in a class by itself. When anyone can spin off a new idea, basically self test on the side, and roll into production, with apparently zero regression testing, all kinds of weird stuff can get rolled out.

I don't have a lot of hope for anything. Heck, in a recent update, somehow a nice new feature, the headlight card pop up on stalk push, got killed. Who knows why, maybe an accident, maybe on purpose?
That's unfortunate to hear talent at Tesla being lost to increasing stress and chaos - you'd think they were over that mountain with getting Model 3 and Y off the ground. I wholly agree that for many employers, in many industries, there is subpar management who forgo any good design thinking (and the integration of it) in favor of rapid, chaotic development cycles that mostly serve their own personal goals. There are very few companies who can really string together coherency because of proactive leaders within the company. Apple is another example of falling under their own weight in software quality: iOS and MacOS have been outpaced by the hardware teams with the M1 chips. The quality of their UI/UX and software across the series of stock apps is surprisingly clunky coming from the golden age of the iLife suite of the 2000's.
 
Thanks for the detailed comparison, you're absolutely right on all of it. I work in software and have had the pleasure of working with some good UI/UX designers. We all agree we should be agile, release features quickly, get feedback from users and adjust. That doesn't mean we shouldn't have a "main plan" to underpin it all, and that's what seems to be forgotten sometimes.
In some cases in software it isn't that dramatic, no lives are at stake. Here, where driving a car is in fact a matter of life and death, good UI/UX design is a lot more important. It's not just about good looks. I hope Tesla are listening.

Maybe you could work for them? :D

Oh: one comment about tire pressures. I've seen countless people unable to get to them in V10, saying they could only spin the car. the spot where you needed to swipe was too small for most. In that sense, the information is more accessible in V11...
 
  • Like
Reactions: Transformer
Thanks for the detailed comparison, you're absolutely right on all of it. I work in software and have had the pleasure of working with some good UI/UX designers. We all agree we should be agile, release features quickly, get feedback from users and adjust. That doesn't mean we shouldn't have a "main plan" to underpin it all, and that's what seems to be forgotten sometimes.
In some cases in software it isn't that dramatic, no lives are at stake. Here, where driving a car is in fact a matter of life and death, good UI/UX design is a lot more important. It's not just about good looks. I hope Tesla are listening.

Maybe you could work for them? :D

Oh: one comment about tire pressures. I've seen countless people unable to get to them in V10, saying they could only spin the car. the spot where you needed to swipe was too small for most. In that sense, the information is more accessible in V11...
Thanks and good point on the swipe area size! I have applied a few times over the years. I don't think I'm what they're looking for.
 
I think your list points out some specific issues, but I'd argue that the problem is much deeper and more general than that. Things that any human should just inherently know from birth:

1. Critical controls should not be blocked by unimportant status graphics. Duh. Like Homelink being blocked by the seatbelt, texts, phone, turn signal, etc. - nearly every status indicator blocks access to the Homelink control which could easily kill a pet or damage a spouse's car. How dumb would someone have to be to not realize that controls are always more important than status indicators? In any system.

2. Don't copy Windows 8. Ever. There should always be just one UI - never 2. Also, the second UI shouldn't be hidden. Oh, and the 2nd UI shouldn't be completely different from the first either. Maybe just not have a 2nd UI? How is anyone supposed to know that the trip meter is in one system and tire pressure in the other?

3. It's a car, so car stuff should exist on the screen. Like the defroster and compass.

4. Software should be tested before pushing it out to millions of users. Someone should have at least tried it in the parking lot or something. How dumb would you have to be to take an intern's first coding attempt and just blindly push it right out to the entire fleet without even looking at it first?
 
This is what happens when you don't have a true HUD with key functionality. They don't necessarily need to put one into the center console like most cars. But I'd argue they should have HUD that puts environment controls, speedometer, audio, map & FSD vision (I think the FSD vision is larger than it needs to be on the screen) into one convenient area when driving. Right now it's like having an iPad that while is easy to navigate in "P" is not what a user needs while on the move.
 

spokey

Active Member
Aug 8, 2020
1,121
685
Flagtown
Welcome to TMC, and thanks. I've been designing hardware that has to interact with software for IDK, 25 ish years, and I've basically come to all the same conclusions. Sadly, I've seen a massive degradation is management's understanding of what it takes to integrate good UX/UI design with hardware functionality, across all industries. Few companies make the space in the schedule anymore to really get it right. Good designers are quitting in the face of the chaos. I know this to be the case at Tesla, as I've known 4 engineers personally who quit, one worked there since the 2000's until last year, the other 2 for shorter periods. All 4 quit in the face of increasing stress and chaos.

Tesla's development strategy is in a class by itself. When anyone can spin off a new idea, basically self test on the side, and roll into production, with apparently zero regression testing, all kinds of weird stuff can get rolled out.

I don't have a lot of hope for anything. Heck, in a recent update, somehow a nice new feature, the headlight card pop up on stalk push, got killed. Who knows why, maybe an accident, maybe on purpose?
We always called it eye-candy

They could solve most of these issues by really allowing user configuration. I would bury the toy box. Have used it once to try the light show thing when it showed up. I would probably put trips & tire pressure back somewhere easy. I liked the cards from prior releases. Put that back and allow me to chose which panels to show and don't restrict to 3.
 
That's another interesting subject. @LightningRider , what is your professional opinion on configurability / customization of the UI? Plenty of people are asking for that. I'm a power user so I typically love configurability but part of me is afraid of the horrors that people could do with it. And I'm not even thinking of lending your car to someone else...although driver profiles probably solves that.
Maybe forcing some essential / security items and leaving the rest configurable? I thought that's what they were doing with the launcher bar but it's just "apps".
 
We always called it eye-candy

They could solve most of these issues by really allowing user configuration. I would bury the toy box. Have used it once to try the light show thing when it showed up. I would probably put trips & tire pressure back somewhere easy. I liked the cards from prior releases. Put that back and allow me to chose which panels to show and don't restrict to 3.
If Tesla really wanted to poll their customers for preferences, they'd do some A-B-C-D testing of the new software, skinning it with v8, v9, v10, v11 aesthetics and hierarchy, and allow a select set of users try them all, and choose the one they like best. Nothing about v11 functionality precludes skinning it with anything.

Ha ha, I kid. They obviously don't have (or prioritize) the bandwidth to even get one UI right, no way they could do 4. Surprising for a company with the resources they supposedly have.
 
  • Like
Reactions: LightningRider

spokey

Active Member
Aug 8, 2020
1,121
685
Flagtown
If Tesla really wanted to poll their customers for preferences, they'd do some A-B-C-D testing of the new software, skinning it with v8, v9, v10, v11 aesthetics and hierarchy, and allow a select set of users try them all, and choose the one they like best. Nothing about v11 functionality precludes skinning it with anything.

Ha ha, I kid. They obviously don't have (or prioritize) the bandwidth to even get one UI right, no way they could do 4. Surprising for a company with the resources they supposedly have.
They don't even need to poll / user testing. Just let us change it around and opt-in to provide data feedback. That might give them a 1,000,000 person focus group.
 
The essential problem is that of an increasingly large software company: there are multiple internal groups of developers and product managers who feel the need to 'make their mark' and do something new for promotion. Microsoft, under Ballmer, was the prime example. For multiple years, the Office group (which earned the most profit), created their own UI design, UI code and UI libraries which weren't part of the rest of Windows OS, which nominally had the responsibility of defining the UI look and underlying technology for all Windows software.

The problem here could even be Elon, as some young developer or product manager caught Elon's attention with some glib clever sounding platitude like "all input is user error" or some such insane bullshit, and Elon lights up and orders them to ship it. The old dude who was previously doing that work thought that was a load of crap, tells people so, and gets fired. New 25 year old prod manager hires 15 interns from Stanford and pushes through all the changes. The life-safety and device driver embedded software developers keep their heads down and promise one another to never talk to Elon.
 
That's another interesting subject. @LightningRider , what is your professional opinion on configurability / customization of the UI? Plenty of people are asking for that. I'm a power user so I typically love configurability but part of me is afraid of the horrors that people could do with it. And I'm not even thinking of lending your car to someone else...although driver profiles probably solves that.
Maybe forcing some essential / security items and leaving the rest configurable? I thought that's what they were doing with the launcher bar but it's just "apps".
Customization can be a good thing. For Tesla it could accomplish several things: users could rearrange their screen (or have multiple setups saved) to suit their preferences: Performance models could have a no map, purists could remove everything but the odometer.

I think customizing has to be balanced with what is the intended purpose behind the product as a whole. Maybe the idea of everyone having a different Tesla experience isn't the same as everyone have a general, good, experience. The latter, V11 aside, may be where Tesla likely continues, though they are making their OS much more flexible as shown in the new Model S and X: the music player that can be dragged around and minimized. Wouldn't be surprised if we see more "preset" configurability like that in future OS releases.
 
@Gauss Guzzler Spot on w/ software testing. I'm constantly amazed Tesla continues this practice at this level of scale. It's beyond "the fans" that own Tesla's now. Busy parents will be nowhere as tolerant as some fans have been for experiencing bugs.

@DrChaos having worked in Tech for a while, and knowing some ex-employees, I realllllllllly hope that was not the case. On reddit, I've previously seen people comment that the previous design lead left. I have no doubt Elon supports the simplifications V11 made - I would have argued otherwise as a design lead for the reasons I originally posted.

@SabrToothSqrl Great Video!! Ironic to see that Apple imo is leading the way in Map design/experience now with their crazy level of detail and choice of camera angle / motions while driving.
 
  • Like
Reactions: SabrToothSqrl

abasile

TSLA shareholder
Supporting Member
In addition to some of the issues mentioned here, it bothers me that the Bluetooth menu is now buried under Controls. As the driver, I sometimes need to switch whose phone is coupled to the car, and this has become more cumbersome. (The person in the passenger seat can’t always handle this, as they aren’t necessarily competent with the current UI, another issue.)
 
  • Like
Reactions: Patrick66

Products we're discussing on TMC...

About Us

Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.

Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


SUPPORT TMC
Top