TSLA Q2 Earnings Report - Quick Take

I’m a Tesla shareholder (5.5% position) and it looks like a solid quarter at first take.

Total revenue grew 47% YoY in Q2 to $24.9B and they re-stated their long term goal of 50% CAGR target

Biggest thing that stands out is the Operating Margins have dropped from 17.2% three quarters ago to 9.6% this quarter. I believe this is mainly due to the price cuts and new product ramps. Economies of scale should help bring this back up & i’m hoping this is the bottom in regards to margins.

Tesla Highlights

9.6% operating margin
$2.4B GAAP operating income
$2.7B GAAP net income
$3.1B non-GAAP net income

Operating cash flow of $3.1B
Free cash flow2 of $1.0B
$0.7B increase in our cash and investments3 QoQ to $23.1B

Cybertruck factory tooling on track; producing RC4 builds
Model Y became the best-selling vehicle globally in Q

2023 Q2 Tesla Quarterly Update

Dave Lee’s take on it:

Key takeaway - Tesla is very resilient, even with current Macro challenges & high interest rates. Things are likely only to get better from here. They will re-stated they will hit $1.8M in sales this year.

-Growth goals haven’t changed,
-Cybertruck to start deliveries by end of the year
-Dojo AI supercomputer ramping
-Innovation & Growth still intact

Tesla’s SURPRISE Q2 Earnings Report (Ep. 734) - YouTube

Solving the money problem commentary on the Cybertruck 7/18

Ford PANICS As Cybertruck Production Begins - YouTube


Their vehicle deliveries were up 10.23 percent sequential and 83.02 percent YoY. That is really impressive. Production was up 85.51 percent. Haven’t listened to the earnings call yet but margins are compressing. They had over 1 billion dollars of FCF.



I think that the ultimate reason for investing in TSLA is robo driving. Auto driving requires AI. AI requires massive data. TSLA already is the world leader in collected useful data for autonomous driving purposes by a factor of 8-9:1 vs all others combined.

More wheels on the ground, even more data for TSLA. Unless autonomous driving altogether fails, TSLA is on the road to dominating the automarket. Complaining about compressing margins is like complaining about AMZN retail margins when AMZN was building AWS.

Would one want AMZN the retailer or AWS? Would one rather have TSLA the fancy car maker or TSLA the dominant autonomous driving player.

That is how I look at it (there is nothing original in what I said, of course). But I don’t look at Tesla as an indefinite holding, rather as a buy-and-forget holding as AMZN, AAPL, MSFT, GOOG in the past.


You can think that, but then you are putting your faith in something where we have no idea how it is going to turn out (how well it will work, what people are willing to pay, how large the market really is, …) and ignoring a whole string of businesses which are much farther along.


??? My biggest holding is TSLA. But it is held on the basis of cars, energy, etc., i.e., areas which are already growing dramatically and where there is a proven market … not based on something that might happen some years from now.


It is already happening, Mercedes-Benz is at level 3. I am not sure what level Tesla pundits claim level 2. That is how technology works, its not bam it’s here, it’s slowly grows. Look at 5G, would you say they have 5G all across the country or would you say it’s at level 2? But everyone investing in Verizon is investing on 5G because that is what they need in order to succeed. Just like TSLA.

If you are investing in TSLA for the car business alone, it is over valued and should be sold. No way is it worth the marketcap it has right now based on the car and energy market. Because the Solar business is pathetic and the car business while doing good, it’s not worth a P/S of 9.7.



Hey, just an FYI, but the last few posts are starting to prove why TSLA discussions were off-topic. No attacking each other or denigrating other people’s reasons for owning a stock.


Whoa DL, I haven’t felt attacked and I am not attacking anyone. We are having a difference of opinion but we both hold shares in Tesla.



On FSD, during the conference call yesterday Musk did say:

And we are already in discussions with – early discussions with a major OEM about using the Tesla FSD

Which implies that at least some in the industry think Tesla’s efforts are bearing fruit. As with all things Tesla, it sometimes can be hard to separate reality from the media (professional and amateur) spin.

Just a few months ago, I don’t think there were many who gave Tesla a better than 10:1 chance of getting its charge port/plug adopted as the US standard. CCS was written into most government (Fed & State) incentives, and only one other very small OEM (Aptera) was supporting it. But, that changed dramatically, and we’ve seen how the dominoes fall once Ford made the move.

FSD might look risky now, but one also might also think about adjacent AI products/services that Tesla can provide in the future. Such as the Tesla Bot, or licensing its “dojo” AI server cloud, or in-vehicle AI chip design, etc. FSD doesn’t itself have to be smash hit on its own for Tesla to do very well.


I don’t think you can “set and forget” tesla but I get your point. When will FSD happen? 1 year? 5 years? 10 years? No one knows. Dave Lee estimates between 1-5 years. The amount of data Tesla will have amassed is expected to increase 20x in a few years so that is encouraging. FSD story could end up similar to charging stations, Tesla is already in early talks with an OEM (Ford?) to license it’s FSD.

Tesla just revealed this SECRET FSD Advantage (Ep. 733) - YouTube

The bottom line for me is Tesla has incredible optionality (storage, bots, solar, etc) and is working on a multi-trillion dollar opportunity in FSD with an almost unassailable lead in that arena. Because Tesla is somewhere between a car company and an AI/software company, the valuation suggests that and many investors are willing to pay that premium. Any bet on Tesla is also a bet on Elon, and in the long run, he usually gets it done.

As for it’s core business of EV’s, Tesla is executing on it’s long term CAGR of 50% revenue growth, Model 3 is the best selling vehicle on the planet, and even with the price cuts (that is what you do when interest rates are high plus it squeezes the competition) it still has industry leading margins. Plus, Cybertruck will ship later this year (already 1.5 million on backorder) and a lower priced vehicle (20Kish) is in the works. Add to this the fact that Tesla charging is becoming the industry standard and there is no doubt they are firing on all cylinders (or is it battery packs?)

First mover in an important industry, large and expanding TAM, incredible optionality, visionary leadership, powerful Brand, $25 billion in cash on hand . . . the motley fool rec writes itself.

Long Tesla (6% position)


Here is what Musk, on the conference call, said about FSD.

**In the long-term, autonomy we think is going to just drive volume through the ceiling next level. And our sort of future robotaxi products – dedicated robotaxi products we think have like quasi-infinite demand. The way we’re going to manufacture robotaxi is, is also itself a revolution. So, it’s revolutionary design made in a revolutionary way. It’ll be by far the highest units per hour of any vehicle production ever. So, very excited about that.

With respect to Autopilot and Dojo, in order to build autonomy, we also need to train our neural net with data from millions of vehicles. The more – I mean, this has been proven over and over again. The more training data you have, the better the results. And, I mean, there are times where we see basically – in a neural net, basically it’s sort of at 1 million training examples, it barely works; at 2 million, it slightly works; at 3 million, it’s like wow, okay, we’re seeing something, but then you get like 10 million training examples, it’s like – it becomes incredible. So, there’s just no substitute for a massive amount of data. And obviously, Tesla has more vehicles on the road that are collecting this data than all of the companies combined by, I think, maybe even an order of magnitude. So, I think we might have 90% of all – or a very big number. So, the success in AI endeavors is a function of talent, sort of unique data and computing resources. And we have outstanding capabilities in all three arenas.**

The more cars, the more data, the faster FSD will come, at least that is what I think Musk is saying.



In my post history on another board (Tesla EV Pricing Strategy - #12 by DSNerd), I talked about how in the Q1 shareholders deck they posted a chart showing exponential cumulative miles driven, and what that could mean for reaching autonomy.

What caught my eye in the Q2 earnings deck is how much the miles driven has accelerated. Look at the y-axes on the 2 charts–they added almost 200 million miles in the past 3 months, which is 1.5x the amount that had been driven in the past 2 years! Based on my rough fit, I expected them to hit 500 million by the end of the year after Q1, and with the new data it appears they’ll likely hit 500 million in Q3 (without accounting for additional acceleration).

This explosion of data is why I think more attention is being given to Dojo and their GPU hardware–it sounds like they’re starting to become compute limited rather than data limited (or at least will soon with the rate of miles driven).

It’s difficult to say when autonomy will be fully achieved, but this kind of exponential scaling of data + compute in tandem is really exciting and what’s ultimately needed to get there IMO.

(Data from the shareholders decks: Tesla Investor Relations)


Well, they are making the claim, but if you look at the limitations, it is a wild exaggeration.


Brad Munchen posted a chart on twitter showing the FSD crashes according to NHTSA, noting crashes were up 83% YoY (https://twitter.com/BradMunchen/status/1682792673491681280?s=20).

Knowing the huge increase in miles driven based on the chart I posted above, I was curious to calculate the rate of accidents per million miles driven for each month. The miles driven are approximate since I’m eye-balling them based on Tesla’s charts, but here’s what the chart roughly looks like.

A quick google search provided this website (https://www.valuepenguin.com/car-accident-statistics
) that shows crashes per million miles driven for each state. It seems like most states show rates in the low 1’s, which means the FSD miles driven have been slightly better than average over the past few months.

Lots of caveats apply: all of these numbers are approximate, the trend looks to have flat-lined recently but is lumpy since it’s month-to-month, FSD includes driver supervision currently, the recent dips are during warm weather months, etc. but the progress looks pretty good overall. Hopefully my last post noting data+compute scaling can continue pushing this rate lower and lower.


New video out from a Shanghai promotion site with a tour of the Tesla GigaFactory there:

Big news is the reveal from a Tesla employee that Tesla has achieved a 40 second per vehicle run off the assembly line. To put that in perspective, half a decade ago Ford was touting their 53 second per vehicle roll off for the F150 ICE version:

Just a couple years ago, VW’s then CEO was pushing his team to get total assembly time down from the then current 30 hours per vehicle to Tesla’s 10 hours per vehicle: Tesla Can Make an Electric Vehicle in 10 Hours, Volkswagen Production and CEO Unable to Keep Up

With this latest news, maybe Tesla has gotten that 10 hour/vehicle assembly down even further.