Will you buy FSD before the $1,000 increase on July 1st?

[Dan203]

Who said indefinitely?

That's the whole point of offering it as a service. We're not talking about loans. This is software as a service. It's all the rage these days and offered by most major software companies. It allows you to offer your product at a much lower upfront cost and in exchange you end up with more customers paying a perpetual monthly fee that actually increases your profits.

I'm sure that them doing this will p*ss off a lot of people who actually paid $5-$8k up front for FSD, but it'll likely result in a bigger income for Tesla.

 

[SammichLover]

That's the whole point of offering it as a service.

No it isn't.

We're not talking about loans.

If you don't include opportunity cost, and %3 is a very reasonable number for that these days, then your economics prof will give you a big fat (F).

It allows you to offer your product at a much lower upfront cost and in exchange you end up with more customers paying a perpetual monthly fee that actually increases your profits.

Except it isn't perpetually. It is indefinitely. Which is a different thing. It lacks the commitemnt of buying into the whole $8K. The kids used to call it renting.

I'm sure that them doing this will p*ss off a lot of people who actually paid $5-$8k up front for FSD, but it'll likely result in a bigger income for Tesla.

Unlikely when you're heavily cannibalize a 30% conversion rate of $8K up front.

 

[davidho83]

That's a good candidate for the Myth thread that's going on here.

Besides your list not being exhaustive, nor the bolded sentence being accurate, Cadillac prices its low rent version of NoAP at $5600. Subjectively not worth it to you? ¯\_(ツ)_/¯ I guess you aren't the market, there's only maybe 1/4 to 1/3 of buyers that are. But cries of FSD as the feature set exists right now, without future updates counted in, being overpriced doesn't square where the automobile market on the whole is. That's why I, having paid $7000 total for FSD, consider it to have been a smokin' deal for me.

P.S. I've always had Autopark and have never used it. Summon might be a "party gimmick" but I was entirely shocked by how much enjoyment I got out of using it a couple times. Of course I got it for "free", I never counted on getting or looked forward to getting it, I'd valued the package without its existence. It just showed up one day, bonus gift, and it blew my mind. So maybe I managed my exceptions differently?

Lol it sure sounds like you don't read many threads on here or Reddit. There are countless posts about how AP/FSD didn't detect a closing lane and the driver had to take action, phantom braking, traffic light control causing people to swerve into weird lanes, etc. There have been numerous times I've let NoA do its thing and it will not be able to process merging or diverging lanes on an on-ramp. It constantly changes lanes if you let it do so without confirmation and will sit for a long time with blinker on waiting for a gap instead of detecting gap before turning on the blinker. If you turn off the lane change without confirmation, it nags me constantly to get out of the passing lane when I'm two lanes from the left or will request to change lanes to the faster lane only to tell me to change back without passing anyone. I've also had it alert me to take control because it is encountering something it may not be able to handle and I drive on major highways in a big city (Dallas) which doesn't have terribly abnormal highways.

From this thread, it is clear that there are two camps, one that wants to support advancing FSD and has money they're willing to spend and the other that is maybe more pragmatic with their money and wants value. Additionally, everyone's driving style and experience is different. I admit that TACC and autosteer are awesome and miles ahead of anyone else but to make a blanket statement that my post is full of myths is uncalled for. Not sure how Cadillac is relevant when the question here was is FSD worth $7k/8k.

I also am aware that FSD is an evolving feature and will change in the future. However, given regulations, I do not see our cars ever being full self driving without constant driver monitoring and input. We will not be able to take our eyes off the road and as such FSD is a misnomer. So, for me, and for many people, shelling out $7-8k for a limited feature set in relation to what is promised isn't worth it. Additionally, I've seen many posts where people are irritated that they bought into the promise of FSD features in 2-3 years and never got those features by the end of their lease/ownership.

Again, don't get me wrong, I think what Tesla is doing is awesome and am glad to be a part of it. I just don't see the worth of $7-8k for a half empty promise.

 

[Twiglett]

I know - it sounds crazy, but I hate people driving past thinking I'm that old dude who insists on doing the speed limit to make point.
Especially when the car is consistently 1-2mph slower than indicated - so I'm doing 29 in a 30 where everyone else is doing 35.
Even worse on the highway.
Several states will pull you over for impeding the traffic flow.
I'm OK with testing and trying out and have been doing that with autopilot all this time, but current traffic light function is borderline dangerous and is uncomfortable to be in car when its working. Its like being driven by a nervous learner driver with bad eyesight.
I usually love getting new features, but this should have stayed with early access for a month or two more.

 

[Dan203]

No it isn't.

If you don't include opportunity cost, and %3 is a very reasonable number for that these days, then your economics prof will give you a big fat (F).

Except it isn't perpetually. It is indefinitely. Which is a different thing. It lacks the commitemnt of buying into the whole $8K. The kids used to call it renting.

Unlikely when you're heavily cannibalize a 30% conversion rate of $8K up front.

You do realize that they specifically said that this was on the roadmap during the Q1 call? This isn't speculation, it's something they are planning to do. The only question is how much are they going to charge for it. I'm pretty sure that if they charge $130-$150/mo it's not going to attract many new users, and they specifically said that the intention of this program was to convert more of those owners. I mean if someone was willing to spend that much they would have just bought it up front and rolled it into the loan. It needs to be significantly cheaper to attract a significant portion of that 70%. But if they can find the sweet spot they will attract enough new users to offset anything they cannibalize from the other group.

 

[SammichLover]

You do realize that they specifically said that this was on the roadmap during the Q1 call? This isn't speculation, it's something they are planning to do. The only question is how much are they going to charge for it.

That's the premise my posts are constructed on. I'm simply pointing out what Tesla is giving up if they were to significantly cannibalize upfront sale of FSD, pricing as your suggestion is aiming at. EDIT: Contrasting with more-so supplementing their sales with subscriptions.

I'm pretty sure that if they charge $130-$150/mo it's not going to attract many new users, and they specifically said that the intention of this program was to convert more of those owners.

They're already getting 30% conversion on a full commitment to buy it outright over 5 years (and having to get credit on the car loan/lease for that extra amount) with it locked into that vehicle, at the equivalent of $125/month.

The benefit of renting is NOT getting something 'cheaper'. The entire premise of the concept is you trade off a higher total price, if renting over an extended period rather than if you simply bought it in full upfront, for the convenience of monthly payments and a lower commitment that you can walk away from much early (and since "repossession" would be very low friction, it is unlikely that Tesla would see a need for a credit check to cover this extra cost). Thus you can raise the per month above that. And Tesla is now upping the ante to $8K, which is easily roughly $145/month equivalent (based on 5 year target, which is a very typical car loan to buy period, so it make sense that people thinking like this and that this would be aimed at would be thinking about it for a period about that long to "own", it'd be a stretch to do a 3 year amotorization).

If you don't grok that you simply don't grok the use of monthly payments to drive extra sales.

 

[Fernand]

On the subject of the auto-wiper and possible coatings, I've been happy with auto mode on most releases. Using ONRW (ONR w/ wax) then a spritz of OptiSeal while drying, same as on the rest of the car, seems to be just enough of a coating.

It's critical to keep the windshield clean for the front cameras, and it affects the wipers' detection too. I can understand why they don't want to make more of an OCD issue here, or imply it's finicky, but IMHO the whole system operates sub-optimally if there's any dirt in the way of any cameras/sensors. A lot of things improve if I always do a pre-flight walk-around and wipe.

Two words on FSD: changing definitions. There are those who are hung up on what it should be, when, and those who use it. It's like the evolving auto modes on high-end cameras: good enough for most, selectively usable by others, and a vortex of resentment for a few.
.

 

[DrtyJrze]

Driving at the speed limit is absolutely a hazard on many roads.

No way someone would pay $150/mo indefinitely when they could just pay $8,000 up front, and roll it into their loan, and have it paid off after 5 years. The whole point of selling it as a service would be to make it a perpetual income stream that the consumer never actually owns. If you price it so high that it's basically the same, or more, then the current offering you're not going to attract more customers to the option.

I can't find the exact number right now, but IIRC during the Q1 call they said the uptake on FSD was less than 30%. Which means the vast majority of buyers aren't buying FSD. And I suspect that number will continue to go down as it gets more expensive. If they go to a software as a service model they'll have to charge significantly less, but they'll likely get more customers overall, evening it out. Plus if they sell as a service they don't have to worry about lawsuits if they under deliver on promises or take longer than expected to develop the technology. And they'll be able to realize the full income immediate, instead of the way they're doing it now where they only realize a portion of the income from FSD sales until they reach certain milestones in functionality.

There is a reason Adobe went from selling a copy of Photoshop for $900, to charging just $20/mo for it. They don't make as much sort term, but they make it more approachable, thus attracting more customers, and making more money long term.

I think if Tesla were to lower the price of the service to something reasonable, like $20-$30/mo, they'd attract a much larger percentage of customers and end up making more money long term then they make now selling it for one big lump sum up front to a small percentage of customers.

Dan, we’ve gone over this on another thread:

FSD Subscription

Mathematically it makes zero sense to charge even $150/month for FSD. $30/month is a meth head dream. If you have time to read through that thread I’d love to hear a logical response to support $30/month.

Spoiler Alert:






There isn’t one.

 

[dc121gw]

Seems like there is a lot of repetition in these kinds of threads.

FSD is a vanity buy at this point. I did it because I bathe in the envy of my enemies.

As long as there are shallow people like me, unmotivated by logic and reason, Tesla is going to keep charging high prices for FSD.

 

[nayr14]

I think "FSD" will be restructured again later this year, similar to the change from EAP->AP/FSD. They have serious challenges they need to face that I don't believe the Model 3's sensor package will fulfill, even things as basic as left and right turns at various intersections will be very tricky to accomplish safely without driver input. We are just beginning to test stop light detection, something that's relatively simple for the cameras.

Like most, I didn't purchase FSD because I didn't want to fork over $7k for summon and auto lane change. I believe the 30% take rate of FSD from the forum spreadsheet is skewed by Tesla enthusiasts, my guess would be closer to 15%. The price is simply too steep for the few extra features owners receive, and owners know they can add it later. Even then, I have no interest in having the car drive itself (beyond AP and lane change). If they want $$$ from subscriptions, they need to restructure of the offering somehow. Unfortunately there aren't that many FSD features...

 

[Lectricity]

FSD is beta at best. I've participated in many beta software tests over the years. They cost me nothing but my time and expertise, and I got a discount on the software when it was released to the public. I can't imagine paying $7K or more to help develop a software service that currently endangers my life when it glitches. I'll change lanes and take offramps on my own, thank you.

 

[SoManyM3s]

Ive hit road debris too many times on Autopilot.

I’ve also been almost screwed on a ending lane as well as car not even attempting to stop for a lanes closed due to an accident with a disabled car in lane - also on autopilot.

I really doubt true FSD will happen in the next 2-3 years.

 

[ElectricIAC]

FSD is beta at best. I've participated in many beta software tests over the years. They cost me nothing but my time and expertise, and I got a discount on the software when it was released to the public. I can't imagine paying $7K or more to help develop a software service that currently endangers my life when it glitches. I'll change lanes and take offramps on my own, thank you.

Like when it panic brakes on the highway because it thinks it needs to stop for a "traffic control"
New stoplight/sign feature needs a toggle to disable on limited access freeways. I know some places have ramp meters but it should be able to tell the difference.

 

[Jedi2155]

FSD when it was $3k on top of EAP? Easy decision for me. FSD now when basic AP is free? Thats a far more questionable one. Since I enjoy testing and have disposable income, I'd probably say I would buy it.

If I was pinching pennies, heck no.

 

[yossi]

So nearly 4 years later, they're going to be back to the price they charged the early suckers for vaporware in 2016.

Yeah... still not convinced.

there is a reason why Elon picked JULY 1st since it is the end of the second QTR so he make announcement 2 months before that pepole will be "scared" and rush to donate 7k to boost the bottom line profits i love Elon but this is not cool why not give it for 1k and when i can have it as a robotaxi charge me 100K for it if i decide to do it????

 

[Knightshade]

there is a reason why Elon picked JULY 1st since it is the end of the second QTR so he make announcement 2 months before that pepole will be "scared" and rush to donate 7k to boost the bottom line profits i love Elon but this is not cool why not give it for 1k and when i can have it as a robotaxi charge me 100K for it if i decide to do it????

FWIW they do not get to recognize all the FSD income as revenue until they actually deliver FSD.

They do get to recognize SOME of it now though since SOME of the promised features (of the new version of FSD) HAVE been delivered... (in fact, 6 out of 7 of them have now in the US.... 5/7 elsewhere)

The unrecognized revenue is actually a liability on their books- since it represents money they took for a product they still owe.

(None of that changes the basic reason July 1 makes sense as a date of course- just that they're not seeing 7k in profit with each purchase right now)

 

[yossi]

FWIW they do not get to recognize all the FSD income as revenue until they actually deliver FSD.

They do get to recognize SOME of it now though since SOME of the promised features (of the new version of FSD) HAVE been delivered... (in fact, 6 out of 7 of them have now in the US.... 5/7 elsewhere)

The unrecognized revenue is actually a liability on their books- since it represents money they took for a product they still owe.

(None of that changes the basic reason July 1 makes sense as a date of course- just that they're not seeing 7k in profit with each purchase right now)[/QUOTE

 

[yossi]

most or all of what you said doesn't make sense it is pure income!!!

 

[DrtyJrze]

Like most, I didn't purchase FSD because I didn't want to fork over $7k for summon and auto lane change. I believe the 30% take rate of FSD from the forum spreadsheet is skewed by Tesla enthusiasts, my guess would be closer to 15%.

So first it was a 40% take rate for FSD, then 30%, now 15%, lol. I love this place.

I can save you the trouble of racing to the bottom. The take rate for FSD will have to be around 4% to even begin thinking about the pennies a day subscription plan so many desperately want being viable. Who thinks only 4% of Tesla drivers have FSD?

As I said before, every feature isn’t for every person, and that is okay. Currently there is no need for Tesla to beg people to MAYBE pay $15/month...sometimes.

They also don’t need you to have FSD to pull any data they want from your car. I’m sure the cameras can still see the green light and know you stepped on the accelerator as confirmation, lol.

So make a real business case or be happy and enjoy your basic Tesla EV.

 

[DBV1]

I am considering purchasing a Tesla 3 and not sure why I would not buy the FSD too? Kind of is one of my points of buying a Tesla is the tech and how it improves. I think I would need FSD to do that and get the most out of what Tesla offers.