I guess because most of their competitors are 100 yrs old? Making cars is really, really hard. Tesla’s first mass produced car was only ten years ago, and their top two models 6 and 3 years ago. The average age of their entire fleet is in the low single digits.
There's a HN meme of sorts that goes like "I can build that same website functionality in a weekend, why does company XYZ employ 5000 people to do it?!"
It's a recurring theme because there's some truth to it. Someone can indeed build the core functionality of that website in a weekend, if they ignore 99% of the difficult work of doing it in production and supporting customers and all the rest of the mundane but important parts.
Building a car isn't that hard really. Lots of people do one-off car build in their garage with basic tools and a shoestring budget. And they can build a fun car which works! Sure, it's not maintainable nor supportable but they built it and it works great.
So while Tesla is well past the guy in a garage stage, they are still at the stage of "building cars is really easy" (if you ignore all the maintainability and supportability and long tail of parts availability and everything else that established manufacturers know how to do).
Honestly, That's not good enough. In the car world you can recruit car engineers to fix simple car production problems. Objectively most of Tesla's issues are due to the poor fit and finish of the cabin and general fitment of things together. IF you ignore the whole electric engine and just treat it like and ice vehicle, Telsa should just be building on the knowledge of others and hiring Toyota engineers to help the nail things like delivery a simple service schedule. Making sure that cars roll of the line with everything tight and having spares ready for consumable parts.
It's not an acceptable excuse to shrug your shoulders and say that they're new at this. The concept of delivering a vehicle isn't new and Telsa has a TON of money.
Change the genre of product to Tech / software engineering. If Telsa was a software company shipping a crap product that broke all the time and then suddenly you found out that they didn't do XYZ of industry standard you'd hold them to account for not taking standard approaches to standard problems.
Tesla produced the same number of cars in 2022 as they did in all years combined from 2012-2020. If you think it's easy to just stand up that level of production capacity, and that they are actually they are just ignoring simple production problems, then you are underestimating the challenge.
Things like panel gaps are not necessarily easy fixes - i.e. what if you know the solution to a 1mm improvement, but can't easily fix it without taking the production line down for x weeks because you need to change the robot arm and conveyor system?
There is a concept in manufacturing called the andon cord (a button / pull that activates a claxon and stops the line). Once the line stops, everyone swarms the problem and works on a fix. Toyota in the 80s pioneered its use and then beat the big 3 at manufacturing quality.
They aren't pulling the andon cord because the panel gap's are within tolerance but could be improved.
Parent post is talking about general fit and finish (i.e. tolerances being wide), try being a production line operative and pulling the andon because you think panel gaps could be decreased.
You are supposed to be pulling it if you notice the wiring loom is broken or something like that, not just fit and finish.
Besides, people are talking in this thread like lean / six sigma are new concepts that won't be already be fully embedded in the existing Tesla plants - this stuff is basic manufacturing which will already be fully embedded, but is just misunderstood by most people.
It seems that one basic misunderstanding is the idea that you could just jump into the deep end and start producing six sigma quality at scale. You can’t. Each sigma represents an order of magnitude of additional work at refining the production process, and that does not happen overnight.
Totally agree - you don't put an andon cord on your production line or 'do six sigma' and suddenly get perfect car quality - it's a slow and incremental process.
I guarantee they wouldn’t shut down the line while they wait for a fix for within-tolerance panel gap improvements - they would work on designing a fix that would be implementable on the line while continuing manufacture. Designing, Testing and Commissioning an improvement will often take many months.
The zero defect thing is fixing issues early rather than fixing them at the end - not shutting down the whole line if you identify that a tolerance can be improved and waiting for the fix.
As I understand it, the parent post is talking about defects that cause actual issues. If that's the case, then it sounds to me like the gaps aren't within tolerance.
Then you take it down and not make a sub-par product. I'm not sure why we all need to ascribe some sort of Messianic vibe to a car company. I'm also not sure I want to bow to the altar of "whip your well-paid employees to sacrifice the best years of their lives to push out meh vehicles" either.
1000% behind the electric push, but dang y'all, it is car company
People don’t realize that even corporations have budgets and have to choose what to invest in. It’s not as if TSLA is paying dividends. All of the money that comes in goes to somewhere.
They chose to invest in certain things (sales, manufacturing ramp up, pickup and semi development, charger network, driverless development etc.) and not in others (panel gaps, fit and finish, etc.)
Easy to decide you disagree but it is working for them.
>In the car world you can recruit car engineers to fix simple car production problems.
I'm no "car engineer", but I doubt most of Tesla's production issues are "simple".
>Change the genre of product to Tech / software engineering. If Telsa was a software company shipping a crap product that broke all the time and then suddenly you found out that they didn't do XYZ of industry standard you'd hold them to account for not taking standard approaches to standard problems.
Software is the last comparison I would have made. First of all Tesla is a software company. Secondly, the software industry is notorious for terrible products gaining massive market share. cough Google Docs/Sheets/Slides.
I don't know why you think Google Docs, Sheets and Slides are terrible products, I personally think they're great and much better than the alternatives that came before them, particularly in the problems they wanted to solve: fully Web-based document creation, editing and storage, and collaboration.
He merely asked why it’s considered a startup. Startup companies take employees from established competitors all the time. Startups can also make products poorly too.
Their valuation is three times larger than Toyota, yet they manufacture only 10% as much cars a year. It’s still deeply rooted in hopes of future expansion and innovations, lack of which they might not survive.
They sold about 1/8 the cars Toyota did in 2022 but they also made about 8x the profit per car. And Toyota has a huge debt load whereas Tesla had money in the bank + was growing + FSD will Soon™ be here (and if/when it does, Tesla's profits will explode) + there's the energy business + there will probably be something like AWS for AI in a year or three.
Funny enough I can comment on this with some data…
I was involved in a study a few years back that looked (in part) at how engineering undergraduates defined entrepreneurship and startup. It fortunately the results from that part ended up on the cutting room floor to focus out main paper on something else…
Generally, to undergraduate engineers, entrepreneurship and startups are defined by the following things:
Being in Silicon Valley/San Francisco
Doing something related to technology
Being financially successful (yes…noting another finding was that they often subconsciously conflated revenue and profit, but only when told the financial data was from a “startup”)
In mentality, I guess. If I was pressed to give a definition right now, I'd say a company stops being a startup when they have self-sustaining revenue without continual outside investments and subsidies propping them up. That hasn't been the case for Tesla the entire 20 years of the company's existence.
Also, valuation is effectively meaningless. That valuation has nearly zero correlation to Tesla's actual value.
tesla is a startup because it allows less strict oversight from the public. “startup” seems to now simply mean “burn cash until market dominance” or “regulations have mot caught up to our industry”. it is not an actual startup, just a court of public opinion one, allowing it to do shady shit under the guise of progress at any cost
Edit: I even missed a zero, it has a 800billion valuation