I am creating this thread to continue the current discussion on FSD payment by subscription. People may also want to discuss the model being used for non FSD payments.
You can install our site as a web app on your iOS device by utilizing the Add to Home Screen feature in Safari. Please see this thread for more details on this.
Note: This feature may not be available in some browsers.
Green has found code in the source that points toward an FSD pay-as-you-go subscription model: Tesla has hidden pay-as-you-go FSD subscription plan in source code - Drive Tesla Canada
Would Tesla be able to recognize subscriptions as revenue immediately?
There are a number of advantages to a subscription model:
This year, as FSD really becomes feature complete and starts to mature, is probably a good time for a move to a subscription model.
- Recurring income
- Makes it less expensive for Tesla owners to trade up to a new Tesla as they do not have to repurchase FSD
- Makes it less expensive for existing Tesla owners that do not have FSD to upgrade to FSD (and hence more likely)
- Makes it less expensive for people that buy a used Tesla to upgrade to FSD (and hence more likely)
- Allows Tesla to have different service charges. In particular to differentiate use of FSD as part of Tesla Network/Robotaxi
- Simplifies options both for a new purchase and for used Teslas
There are probably factors I'm missing as an outsider, but when the FSD software is safer than a human and enables robotaxis, $1B immediately is much more useful than $2B over 5 or 10 years. To understand why, one needs to appreciate the economics of a robotaxi:
These numbers are on the conservative side. Even $1 per mile would significantly undercut competition.
- Let's assume $30-35k up front costs (should be lower long term, but base M3/Y and base Cybertruck cost about this much to manufacture).
- Say a cost to operate of $0.20 per mile (Elon quoted $0.18 in his presentation, could be much less long term).
- Assume $0.70 price per mile compared to the $2-3 that other ride hailing services charge.
- At 50% useful miles, that'd be $0.35 revenue per mile.
- This means $0.15 profit per mile.
- Assume a robotaxi operates 100 hours per week (should be 24/7 minus charging and maintenance time initially when there's limited supply, but could still be far more than 100 hours long term, if taxis transport cargo during off-peak hours).
- If the robotaxi travels at ~25 mph (similar to Uber), that'd be 2,500 miles per week.
- That means a robotaxi would earn $375 per week, and ~$20,000 per year.
Nonetheless, the initial $30-35k investment pays for itself within 2 years, and then could generate as much as a 50% or more ROI per year.
This is why from my perspective I think it makes sense for Tesla to prioritize up-front cash, so that it can buy as many of its own vehicles as possible once FSD is safer than a human. However, there could be other variables that favor the subscription model, that as an outsider i am simply not aware of.
Operating one's own fleet of robotaxis is the mother of all recurring revenue
Once FSD is safer than a human, FSD attach rate should be 100%*, and it'll make sense for Tesla to get this cash up front to invest in its own fleet of robotaxis.
*People or investors will line up to buy a $40k EV + $20k or $30k FSD option. $60k or $70k for a robotaxi that can make you $200k or more over a 10 year period is a pretty damn good investment.
Funny, I've made several posts about a subscription based model for FSD in the past. Don't think I've ever had anyone agreeing with that view with a supportive post. Plenty of folks disagreeing though.
"News" to indicate Tesla themselves might think it's a way forward and suddenly lots of posters think it's a good idea. Welcome aboard folks.
Normally I see subscriptions as bad for the consumer, great for the company (see Adobe stock the last decade) but in this case I think it would be beneficially for all parts. As long as the price is set sensibly of course.
But it’s not going to be 7k. It’s 7k today. Elon already said the price is going up again in June? July?
At what price does it become more than the average person will pay upfront? I think we’re already pretty close to that price now.
I wonder if once Tesla Network comes online if Tesla will even offer none network owners the chance to buy it for private use? Or even if it’ll be ‘affordable’ to buy for private use. In which a subscription works.
I’m excited to see where it goes and even more delighted I preordered it on my CYBRTRCK.
To me it makes a lot more sense for Tesla to tie the ownership of the FSD license to the owner and not the car. It provides an incentive for a Tesla owner to stay with Tesla for any new car they buy going forward as they’ve invested thousands of dollars that they would lose by switching brands (since its not tied to the value of the car). When Tesla then sells the used car, they can sell a license to the new owner (probably at a higher price if it’s gone up again), which will in turn incentivize them to stay with Tesla.
OMG. Is there not a "I've had a good idea about how Tesla should sell FSD" thread.
But that isn't what you wrote:
Then you are saying the license is linked to the car, not a person. So if the person owning the car dies, the license disappears from the car even though the car still stays in the household? (How does that work when three people own the car?)
I would like to have the FSD stay with the owner while owning a Tesla. Have it transfer to a new Tesla if same owner. If owner does not buy a new Tesla and does not own a Tesla for 6 months, they lose the license. When selling, new owner does NOT get the FSD license.
This would incentivize people to stick with Tesla and in upgrading.
It’s true that we would lose the cost of FSD when selling and not buying new, but oh well. Many people trade in or sell to dealerships anyway.
Tesla would make a killing with this. I know I’d be more apt to upgrade. And Tesla gets the full FSD cost from each person at one point.
Or just switch everyone to FSD SaaS Model at some point.