Tesla Q3 Call With Financial Analysts – Everything We Know



It’s time to talk Tesla

It’s quarterly financial reporting time again for Tesla. Its third-quarter results reveal an even larger than expected largest-loss-to-date and the pushing back of the Model 3 production timeline by about three months, but also record sales of its Model S and X. With analysts keen to gain insight into a company going through a seemingly turbulent time — besides stumbling on Model 3 production ramp up, it had mass firings of workers from both its automotive and former SolarCity staffs, a delay in its Tesla Semi reveal, and a $500 million raising of its warehouse agreement, among other things — we expected a lot of questions surrounding these issues.

Tesla Model 3

Blue Tesla Model 3

Musk, perhaps a little under the weather and, no doubt, exhausted from trying to get the Model 3 production train back on track, kicked off the call by talking up Model S and X production and the company’s incredible growth over the past five years, predicting that it would, in the coming five years, grow by an order of magnitude from today’s position. Maybe two. Then, he turned to excoriating the media about its coverage of the recent mass firings. He was noticeably perturbed. Actually, scratch that. The man was pissed.

Apparently he took great umbrage that there was speculation in the press about Model 3 production and other aspects of the company’s affairs after, as we mentioned, hundreds were fired from Tesla operations and SolarCity. Of course, the company itself was quite opaque with regards to this issue, sticking with the performance reviews as the reason for letting so many go. When reports from former employees of SolarCity arose claiming no performance review had been held there, the company remained silent. No doubt frustration flows in both directions.

He then went into some depth on the battery bottleneck that’s been a pain in his own neck for the past month or so. To underline the issue and the extent to which they were dealing with them, it was mentioned that the call was being held from the Gigafactory outside of Reno, Nevada. Indeed, we are well aware that Musk has been spending a lot of time there, as he posted a video of himself and others drinking and roasting s’mores over an open bonfire on the roof of the facility.

Explaining the battery bottleneck, Musk said manufacturing of the modules that go inside the packs takes place in four separate “zones.” Two of those were fine, another not so great, and one, well, according to the entrepreneur, the subcontractor dropped the ball and then didn’t tell them the ball was dropped. Apparently this resulted in the need to rewrite a lot of software from scratch and make electromechanical changes to the design. The good news from all of that is that, according to Musk,  the reconfigured zones are now “three times more effective.” To his credit, he shoulders the ultimate blame for the failure on himself and Tesla, since they were the ones who chose the supplier.

Getting into the call, we have to say we were a little disappointed in some of the questions. “What production level of hell, exactly, are you at right now, given that the 9th level is the worst” (we’re paraphrasing a little), is not extremely helpful nor does it lead to insightful answers. In case you wondering, though, Musk says they are in the eighth level, getting ready to leave for the 7th.

Tinted Model 3

Here are a few further bullet points of what we learned.

  • Model 3 production should reach 5,000 per week by Q1 2018, sometime in March.
  • Model 3 is a 10-year project. This might also indicate how long the S and X might last before a re-do.
  • Musk now expects Model 3 production to be in the thousands per week by the end of the year.
  • Autopilot equal to a human’s ability is achievable with current hardware. If required to achieve higher level from regulatory standpoint, hardware (seems to refer mainly to computer processor) will be replaced for owners who’ve already paid for full autonomy.
  • CFO Deepak Ahuja maintains that cash flow will improve significantly with the increased production of Model 3 and capital expenses are being carefully managed. Interestingly, he seemed to imply a slowing of service center and Supercharger rollout in order to keep expenses under control.
  • Gross margins on Model S and X were down due to the sale of inventory cars, as well as the marking down of some 90 kWh cars which they wanted to discontinue. They are expected to return to the mid-twenty percent level.
  • Tesla is trying to differentiate the Model S from the Model 3 by offering more value in the S. Therefore, they all come with air suspension now, for example.
  • Despite a decrease in Model S and X production —  from 2,000 per week to 1,800 — sales are expected to be higher, since they still have some inventory vehicles to move.
  • Tesla is very short staffed right now. (Surprise!) Third shift S and X workers are being moved to Model 3.
  • Musk says next quarter they’ll be able to give more precise information on production expectations. Perhaps some clarity
  • Musk says, “Don’t set your watch by this,” but China factory should start in about three years and produce two hundred thousand  or more Model 3 and Model Y vehicles for that market and nearby regions. Possibly some Model S and X as well.
  • Model S and X quality continues to improve.
  • CPO sales are expected to reach $1 billion for 2017.
  • Musk says they are asking suppliers for faster robots. says they should need a strobe light to see them, that air friction should be factoring into their speed ability. He then hints they may be designing their own robots internally.
  • Manufacturing in China is important for lowering cost, making vehicles more affordable to Chinese consumers. It not related to material supplies for permanent magnet inside Model 3 motors.
  • Expect conventional solar sales to increase as product is brought into stores.
  • Solar Roof is still undergoing pilot engineering, and is installed on more than ten houses. Expect Buffalo factory to begin production by the end of the year.
  • Australia energy storage project appears to be on target.
  • Energy storage battery production is separate from automotive and is doing well. There may be some constraints in the longer term future, due to availability of lithium hydroxide, nickel, separator, and even electrolyte. (They downplayed cobalt constraints.)

And there you have it, the financial call in a slightly large nutshell. Some issues remained unaddressed, of course. The Tesla Semi and surrounding business plan went unmentioned, as did other future product such as the Model Y (outside of being produced in China for that market), and the Tesla Pickup.

Categories: Tesla

Tags: ,

Leave a Reply

45 Comments on "Tesla Q3 Call With Financial Analysts – Everything We Know"

newest oldest most voted

Excellent summary – Thank you.

Given that management is incompetent (see http://www.hybridcars.com/model-3-bottleneck-blamed-on-chaos-and-incompetence-at-tesla-gigafactory/ for the latest) and is manifestly untruthful (or at least criminally naive) in all of its public statements, I’m surprised you give the conference call PR any credence at all.

I would prefer to see an independent analysis of Tesla’s finances and operations.

There is going to be some more discovery coming soon, from some independent analysts. I just hope the Tesla longs hang tough, through what will be about 6 months of Model 3 “production (ramp up) hell.

That insider supposedly works for PANASONIC, not Tesla. And he’s talking about PANASONIC’s local hires from Nevada, and not their Japan management.

Sorry you can’t tell the difference between Panasonic and Tesla.

Nope. If you read his Reddit posts, he makes it clear that he does not work for Panasonic, but for another outside contractor, working within the facility.

He is a contractor for (works for) PANASONIC, not Tesla, and everything he talks about management is PANASONIC management, not Tesla.

“I would prefer to see an independent analysis of Tesla’s finances and operations.”

No, apparently you prefer to believe anonymous posts to Reddit from an attention seeker who claims to be a disgruntled bottom-tier manager working for Panasonic.

“I’m surprised you give the conference call PR any credence at all.”

I’m surprised that InsideEVs continues to allow obvious serial FUDsters like you to continue to post Tesla hater comments.

They’re not anonymous posts to reddit, they are pseudonymous (just like your posts) to electrek.

I tried to trace claims in the Electrek articles back to their source, and what I found was one single guy posting anonymously to Reddit. If you have any facts or authoritative claims that Electrek has another source, then please do post that info here. I certainly am not the first person posting comments to InsideEVs to suggest that Electrek does not exactly have high journalistic standards. Of course, I may be wrong on that point, but it would surprise me if I am. The dangers of relying on a single source for a news report has been well established historically. Anyone who has read the nonfiction All the President’s Men, or has even seen the movie based on that book, should be well aware of the dangers of relying on a single source for investigative journalism. Jay Cole recently wrote: …we don’t give “two figs” what Tesla’s estimate is, we just report from the data we see in the US (and each month we generally get 20-30 unique data points from various differnt sources to go over…and we work 7am to 2am (ET) each and everyday of the month breathing only EVs) Now that is how real investigative journalists… Read more »

Well Jay, 4e has certainly thrown down the gauntlet against InsideEvs and your judgment in publishing this.

The trolling by Shorters including 4E is getting to the ridiculous level.

I don’t think we’ve seen anything, yet. Honestly. Tesla shorts are bi-partisan. Any investor worth their salt knows WSJ narrates negative. It’s when Bloomberg jumps on with Reuters, Seeking Morons, etc, etc, that sentiment breaks sentiment the other way. It’s not the business model that matters, right now. Its a solid plan to remain liquid, which they made no bones about prioritizing on the call.

I’d prefer you got back under your bridge!

“CFO Deepak Ahuja maintains that cash flow will improve significantly with the increased production of Model 3 and capital expenses are being carefully managed. Interestingly, he seemed to imply a slowing of service center and Supercharger rollout in order to keep expenses under control.”

Well, if they are going to take 3 months longer to get Model 3’s on the road, it only makes sense that they would delay rolling out all the superchargers and service centers by 3 months too. Seems like a no brainer.

Didn’t you call me FUDster, when I stated this?

So the biggest production bottleneck is battery module welding. Odd that its only 1 of 4 and that they had diffetent contractors for each one…

This is based on … “says Jake”?

Correct me if I’m wrong, but the source of the claim that the bottleneck was due to problems with welding on the TM3 battery pack is one single attention-seeker posting anonymously to Reddit; a single anonymous source who has had no verification of any of his claims. This should be held up as a shining example of how “fake news” is so easily and quickly spread on the internet. Of course, I can’t be sure that everything the attention hound posted is B.S. …but without any corroborating report or supporting facts or figures from any other source to support his claims, it’s wise to consider everything he claims as nothing more than wholly unsubstantiated rumor, and possibly nothing more than what used to be called “tall tales” before the electronic age. Now, that’s not to say we should accept everything that Elon says uncritically. Elon has a motive to put the best positive spin on bad news, and to downplay anything that looks like incompetence at the senior executive level of Tesla Inc. But Elon is constrained by SEC rules to not make false statements materially affecting his publicly owned business. As an example, when Elon recently misstated the number… Read more »

~700/~33000 were fired. ~2%

Tesla has over 1900 job openings world wide. These were not layoffs. They were terminations for lack of performance.

After performance reviews. What is opaque?

Fired people make all kinds of crazy accusations, especially when they hire ambulance chasers to shake down companies for money.

Sexism,racism,ageism any kind of ism.

Yeah, the Tesla haters really worked overtime in managing to portray firing about 2% of the workforce as a mass layoff, or as if it indicated Tesla was in financial trouble.

It just goes to show that if you repeat a Big Lie often enough, people will start to believe it.

However, in my opinion the most absurd claim that came out of that was the claim that people were fired for being pro-union. Well, given that there was a vote at the Fremont assembly plant over unionization within the past year — a vote which failed — it would be remarkable if there were not some pro-union folks among those who were fired.

Should Tesla have gone out of its way to carefully select, among those fired for poor work performance, only those who had voted against unionization? That’s what those making that charge seem to be suggesting!

It doesn’t take any special trickery. California law classifies a layoff of that size as a mass layoff.

So the terminology is correct.

That is, if this is indeed a layoff. Tesla says it isn’t.

I never used the term “layoffs,” as that has a specific meaning and it doesn’t cover what happened here.

Opaque is when we have a public statement from the company about reasons for a mass firing, (only after being asked by media) but when the press brings up seeming inconsistencies, they don’t engage, but say, “see our original statement,” which is being called into question.

They also refused to say approximately how many workers were affected or any number of other issues.

I know Elon is wary of putting out too much information and having it misconstrued, but a similar thing also happens when you are opaque — people start building theories as to why certain things happen, and are as easily wrong as right

Not sure what is actually transpiring, but I don’t see why there should still be a software issue with the battery at this late date.

Doesn’t matter to me, but seems like people are paying more than the usual attention here.

The modules being made from different companies I’d really be curious to know if there is a good reason for that.

There is a whif of emergency, last minute engineering changes in the answers being given. I’ve repeatedly stated Mr. Musk seems the happiest when there is a Crisis, as for example his statements of ‘wake me at 3 am if necessary for the roadster’, and ‘Camp out with a sleeping bag at the end of the assembly line, if either of those actions would do any good. Perhaps this attitude is why trivialities like Fit and Finish, or repairability of collisioned Teslas, are ignored .

He should be very happy now.

“Not sure what is actually transpiring, but I don’t see why there should still be a software issue with the battery at this late date.”

That part of the article isn’t very clear, but I think what was meant was that the software for running the production robots needed to be rewritten, and certain physical (“electromechanical”) changes needed to be made to the assembly line setup in that “module” of the Gigafactory’s pack assembly production line(s).

I doubt that changes needed to be made to the design of some (but not others???) of the modules making up the Model 3’s battery pack, which admittedly is how it came across when I first read it, Bill. It’s certainly confusing to see the term “module” used in this context. When talking about EV battery packs, “module” usually refers to a group of battery cells, a division inside the pack. I had to go back and re-read that section of the article before I realized there was a more reasonable interpretation.

Strange to see that in an article by this writer. His articles are usually much more clearly written than that.

This is what I understand too. The programming is for the robots, not for the car.


I gotta give Domenick lots of credit on his writing. He basically live-blogged the call and then turned it into a story very quickly. I listened to the same call and took my own notes. And before I could get them all organized to post on the last story, he had this story posted. And I type at a speed that I am told is best described as a “cacophonous blur”. Two big thumbs up for the fast turnaround.

Yes, thank you Nix. Yes, I should have given Domenick more credit here. This was definitely a case where the reporter is obligated to get the article done as fast as possible, not to check it to make sure every “i” is dotted and every “t” is crossed.

Well done, Domenick!

Apologies if things weren’t more clear. I was trying to get the piece out and relied on memory and notes instead of re-listening to the call. Truthfully, I didn’t quite understand exactly what the problem was even after listening to Elon’s explanation. Without seeing the the actual facility and process in person, we can only go by the explanation, which was understandably brief. Here is a transcript of that portion of the call. Maybe re-reading it will help. Certainly, that’s whats I’m about to do a few times. “The primary production constraint really, by far, is in battery module assembly. So a little bit of a deep dive on that. There are four zones to module manufacturing it goes through four major production zones. The zones three and four are in good shape, zones one and two are not. Zone two in particular, we had a subcontractor, a systems integration subcontractor, that unfortunately really dropped the ball, and we did not realize the degree to which the ball was dropped until quite recently, and this is a very complex manufacturing area. We had to rewrite all of the software from scratch, and redo many of the mechanical and electrical elements… Read more »

Perhaps the author edited the article since your post, but it’s pretty clear.

Each battery pack assembly line has four “zones” (i.e. four distinct stages of assembly). The subcontractor or one of the zones “dropped the ball” but kept Tesla in the dark. Once they realized how badly the dumb, evil subcontractor had screwed up, Tesla stepped in and heroically re-wrote 30 man-years of robot control software in 4 weeks. They also made some “electromechanical” changes to the assembly equipment. Now battery assembly runs 3x faster than originally spec’d!

Solar roof has been installed on a whole 10 houses so far and is still in pilot engineering…. if their past is any indication their actual semi roll out where actual people can actually buy it in mass it is gonna be a long ways off never mind the Y.

It’s crap like this that really annoys people with Tesla. I mean Apple does the same thing to a degree with their products but at least they deliver a ton of them at launch and then catch up with demand in a couple months.

No, that’s what makes Tesla exciting. They share a lot about future plans and product, in a messy way.
Apple is the opposite. They tightly lock their communication and only talk about new products when mass production already started.

DJ said:

“…if their past is any indication their actual semi roll out where actual people can actually buy it in mass it is gonna be a long ways off…”

I’m beginning to feel like Cassandra in pointing out that the Tesla BEV semi tractor will be only a prototype or technology demonstrator. There is absolutely no sign that Tesla is moving toward putting a truck into production. The fact that they recently delayed the BEV semi demo, apparently to help with hurricane relief in Puerto Rico, is another sign the BEV semi tractor is just a back-burner project for Tesla.

Whether or not Tesla moves to put a BEV semi into production is going to depend a great deal on how much or how little interest is shown towards the prototype by actual trucking fleet operators. Furthermore, if Tesla does move to put that into production, it will be years off. I seriously doubt Tesla would devote much of their resources towards putting a BEV semi into production until after Model Y production is well along, at least.

Production is difficult.

It is all about solving a lot of problems in a short period of time.

Did work with model S and X and will be so with model 3 also.

Delays is part of the game.

“Musk says they are asking suppliers for faster robots. says they should need a strobe light to see them, that air friction should be factoring into their speed ability. He then hints they may be designing their own robots internally.”

Seriously, more crazy claims.. Like Tesla doesn’t have enough issues on its hand. Somehow Toyota can build 400K Camry without those “crazy robots”. Oh, EVs are simpler to build, right?

I won’t cancel my reservation, but seriously, Elon needs to stop the spinning.

As silly as the statement is, the really silly part is that there should be no need to argue about the top speed of the robots at this point.

The supplier would have given them a spec on how quickly the robots could accomplish their jobs before the contract was even approved. If that was fast enough there is no real reason to argue about it anymore. They just need to deliver on the promised speed.

And Tesla would have then made their plans around robots of that speed. Their plans for meeting their production goals .Their plans for sizing their plants and lines.

If the contractor felt they could satisfy the contract with robots that were visible to the human eye while moving then that’s plenty good enough and no need to complain that the robots should be faster. Meet the goals you laid out first and then speak of the idea of robots which make small sonic booms later.

Yeah, I think Musk would be pretty happy if the machines were simply producing parts at their contracted rates.

Classic case of trying to run before you walk.

Building 400 k Camrys is not impossible with slow robots. You just need a whole lot more in separate production lines, or longer production lines.

A lot of mathematicians and software engineers are paid serious money to optimize production on existing manufacturing lines (all over the world, not only automotive) that involve multi-axis robot arms. Even fractions of a second saved per move by changing the way / speed of moving to target, can significantly improve througput and / or energy consumption of such a machine, thus reducing cost.

A supplier for manufacturing equipment doing a half assed job is not uncommon, perhaps said supplier has even some incentive (big ICE business?) in delaying model 3? Wouldn’t be surprised if the ICE cartel would be behind this, after Tesla took over Grohmann and canceled the future ICE line contracts.

22 ammo production by Federal is 8 million per day. That’s about 93 per second. Even at such level of production, 22 still cost almost double pre-Obama days (though getting better). Maybe Musk wants to produce cars just as quick. If he did, most people who want Tesla 3 would qualify for Tax credit.

To give positive spin on his comment, maybe that’s what he has in mind: quicker production so that more cultists can enjoy the tax credit.

Before that plant was shut down by Toyota and GM, it was able to produce 400k vehicles a year, so I’m calling b.s. tesla should by now know how to mass produce. It’s thier forth vehicle in production

If there are 4 zones to making a battery pack, how did Tesla not know one of them wasn’t working long before it became a problem?

They should have verified the pack production process months ago, before final assembly on Model 3s started. You can’t greenlight final assembly of salable models if you haven’t verified that production of subassemblies is working. And it’s not like they’d even have to make a vendor site visit, this work is done under their own roof.

Months before you should have watched some being produced, measured the resulting products for tolerances, tested them for their construction quality, etc. They should have done all of that. To find out later that the pack production wasn’t working right months later (even 8 weeks ago) just seems like someone missed a trick.

There do seem to be some major inconsistencies in what Tesla (or Elon) is claiming. On the one hand, Tesla/Elon claimed some months or years ago that they sent out Tesla reps to work right on the floor at the manufacturing site for various vendors for Model 3 parts, to ensure there would not be the sort of delays seen in Model X production. But now, Elon is claiming that they didn’t realize there was a major failure — I’d call this more than just a “bottleneck” — with battery pack assembly, because one of four vendors for setting up the assembly line wasn’t able to deliver, but lied to Tesla about not being able to? Where was the close oversight that Tesla/Elon talked about? Are we actually supposed to believe that Tesla used its reps to ensure smooth production at many or most of its vendors, yet did not exercise any oversight of setting up the battery pack production lines at Gigafactory One? I remain skeptical of the claims made by that anonymous supposed whistle-blower posting tell-all posts to Reddit about delays at Gigafactory One. As I’ve said elsewhere, we need to be quite skeptical of supposed insider info… Read more »
I wasn’t really going by the “leaker” for information on this part of the problem. I think the openings he exposes are in other areas which, if true (hard to trust him as a single unknown source) would be even more significant. For example, if he says that Panasonic can’t even make cells reliably in the capacity that they need to make. If that’s the case and they have a yield problem for their cells then that will hurt production for many months. Oops, I shouldn’t go off on a tangent because I was trying to say I didn’t regard this fellow for this particular issue. So let me get back on topic. I was really going by what Tesla said caused their production problems for this particular issue. And I do appreciate your considered input here. You raise interesting questions and settle some points. Your concerns on this front, even if not identical to mine, do mirror my concerns about how this could go wrong in this particular way. I hope they get this settled soon and I hope that they have (even if only belatedly) gone through this check process on enough of their other suppliers now that… Read more »

According to the conference call, the problem was with the inability to scale, not the inability to produce a part at all.

What worked for 30 or 100 cars a month, fell flat on its face when trying to ramp up to 500 a week, like when they tested their other parts of production.

It is clear from the call that at some point Tesla did a velocity test on all of their subsystems, and most of them passed at at least 500 per week volume. This one failed.

Apparently the subcontractor knew it would likely fail, but said nothing and only showed Tesla that it was successful at much slower rates. They talk about this in the conference call. Hard to quote verbatim from memory, so this is best effort paraphrasing. I suggest listening to the call.

I do understand the value of the information in the call. But I don’t really have time to listen to the call. I’ve found if I wait a day or two there will be a transcription. I’ll read that as I have in the past. Until then certainly there are gaps in how up to date I can be. While I do appreciate you explaining the issue more finely I still don’t think it excuses what happened. And I’m not sure you disagree because you say “at some point”. What I’m saying is you shouldn’t be failing any kind of production test so late in the process. If you know you need a certain level of output by December to make your 5K targets you should be testing and finding this out in June or April, not September. You should be asking your supplier to make parts at a high rate of speed many months ahead of time. And you watch them do it. Thus you find out they will fail at high velocity much earlier and you can try to rectify it or lower your estimates much earlier. Tesla either didn’t bother to check up on this supplier early… Read more »

Transcript is now up. You can read it 5x faster than listening to Musk stammering and yammering.

“you shouldn’t be failing any kind of production test so late in the process.”

That’s the thing, it’s not late in the process. Carmakers usually test and debug assembly cells and even large fractions of the line at full speed offsite before ever installing them on the factory floor. There’s no “S curve”, you just install the fully tested equipment, spend a few weeks testing the integration of the different sections, and flip the switch.

Tesla had vendors send machinery as soon as it was built. They are debugging each piece on-site. They’re making cars while they debug, and selling these cars to employees. They (optimistically) hoped this debugging process would take 5-6 months. They’re four months and still where they thought they’d be after one month.

So if you take the 30″ launch” EVs from thier launch event from 220 how many evs they really produce in the three months

“Speed up the Robots to Magician Speed”

(Not a priority that the robots at that speed cannot:

1). Be long-lasting prior to a breakdown.
2). Accurately position the parts to be installed, nor accurately torque them.
3). Not interfere with other simultaneous processes ongoing – nor be warranted by Fanuc or whomever they collapse.)

So you forget about any Fit or Finish issues.
If the batteries are to be manufactured this sloppily, imagine the rest of the car.

I assume Mr. Musk must think people aren’t really listening. If they were, I think it would be hard not to crack a smile.

I’m sure some are going to be upset at my comment – but hey, the Magician Speed thing ain’t my Idea!