Product Launch

Oct 25, 2019

Hi, this is Elon. Welcome to the Teva SolarGlass unveil, if you will. We'll I'll start a very, very brief introduction and then go into questions. The intent behind the sort of solar glass roof is that we can make roofs come alive, that you can have a live roof instead of a dead roof. If all these roofs out there are just gathering sunlight but not doing anything with it. And our intent is that you can convert that sunlight into energy. And I think in the future, it will be odd for roofs to be dormant or dead or not gather energy. And you have to say like what is the future that you want? And the future that I think we all want, most of us, is a future where you look around at neighborhood and the roofs are all gathering energy, they're all doing something useful, and they look beautiful. They're very robust and resilient. And they're preparing the houses that they shelter. I think this is what we will want to say. 20 years from now, look around the neighborhood and that's how it is. Now the solar glass roof is not going to make financial sense for somebody who has a relatively new roof because this is itself a roof that has integrated solar power generation. So it has, therefore, the cost of roofing a house in addition to the cost of solar panels or solar cells. However, we've been able to achieve with Version 3 a cost a price point that is less than what the average roof costs plus solar panels. So if you're looking at 2 options, one is you need a roof or even need a new roof or you need to reroute your house, but you look at the cost of that, the average cost of that and then the cost of adding solar panels to that roof versus the cost of Tesla Solar Glass, which is an integrated roof plus solar panel or solar cell, that the Tesla solar glass roof will cost less. This is, to the best of our knowledge, accurate. And we'll obviously see now that it's debuted whether this is correct. I mean, it's not going to be correct everywhere because there are going to be times when there's a very cheap roof and a very cheap solar panel and there's maybe some corner case where it's not where it is cheaper. But I think maybe 80% of the time or more of the Tesla solar car roof should make the most economic sense and look the best and be the longest lasting. That's our intent. And it's been quite hard to get to this point. This is quite a difficult product because roofs have to last for a long time. And then when you add electrification to the roof, that's obviously a fair bit of complexity because they have wires and you need to make sure that it's all going to be safe and not cause any risk to the house, and also that it will last for many decades. We want this to last for something on the order of 30 years or more. So it's not easy to do accelerated live testing because you want to say, okay, well, how do we, in the course of, say, 6 months, and try to accelerate the life of the roof such that we know what it will be like in 30 years. This is quite hard. So the system has to be extremely robust. And then figuring out how to install it effectively is very nontrivial. And we're actually going to have like kind of install a thons where we've got we have the snub at our near our Fremont factory, 2 houses that are identical and we have the 2 competing teams to say who can install the solar glass roof the most efficiently. And then we're going to try to get like some of the best roofing teams in the country to also give it a shot. So it's not just Tesla internal. And actually, I should say that the solar glass roof, while we were going to start out just having Tesla and Solars do it, this is something that we intend to open up for roofing contractors in general. So, unlike vehicle sales, which only go through Tesla stores or online, And that's because we sort of we don't want to get trapped in like franchise law and there's like a lot of complexities in automotive distribution that if we kind of break precedent there, it's highly problematic potentially in many places. But for the Tesla Solar Grass roof, I think it's a different situation. You don't have like local sales monopolies. It's a highly competitive market, and I think it's going to make sense, therefore, to have like Tesla certified installers who can then install the solar glass roof. And I think that will be a very powerful driver of demand and allow us to grow much faster than we could otherwise grow. And this is all going to be produced at the Giga New York, our factory in Buffalo, where people have been working really hard. And I'd just like to say, no appreciation for the team there. They've really been putting in a huge effort to ramp up production of the solar glass roof. And I was pretty impressed when I visited there earlier this year. That was like, okay, it seems to be like a lot of 100 people here. It seems like a good situation. I think we'll definitely make New York proud about that factory. It looks going to be great. So, yes, it's a really cool product. And like I said, in a nutshell, if you are getting a new roof or if you're reroofing or getting new roof, those are like I feel quite confident this is a smart move. And then but if you have it, if your roof is already new, then it will not be the smart move financially. You may still want to do it, but it's going to be financially punitive to do it on an existing roof that still has a long way to go. And if you say you have an existing roof and that has a lot of life left on it, then we would recommend the Tesla solar panels, like the just get solar retrofit is what we call it. And that's the right move. If your roof still has like, say, 10 years of life or more and then Tesla Solar Glass if it's a new roof or reroofing. All right, and with that, we'll move to the questions. I'll just repeat the questions and then do you want us to answer them? Thank you. We will start the call with the most upvoted questions related to Solar Roof from the retail investors on say.com. Justin asked, what changes has Tesla implemented V1 to V2 and now V3? And what has been their impact on cost or performance? I mean, there's quite a few changes. So I have Drew and Kunal here. Guys, do you want to sort of address some of the changes that we've made? A lot. Yes. I mean, we really wanted to achieve all the goals that Elon just outlaid there, and that involves assessing cost, beauty, installability, manufacturability across all axes. So we've just to run through some of the key points there, we've increased the size of the tile, increased the power density of the tile, dramatically reduce the number of parts and sub assemblies in the tile by more than half. All of this goes in the direction of lower costs and easier manufacturability. We also changed some of the materials in use, changed the methods that we're using to the technology that we're using to achieve the hidden solar cells to something that's more scalable. We I should say, like, this is one of the hardest things about the solar glass roof is not seeing the solar cells beneath the glass because that's kind of off putting to see the cells. And then the hard part is you want the photons from the sun to get to the solar cells but not create something which is aesthetically unappealing. That's quite a hard challenge. Yes. Solar cells are optically isotropic, meaning it can look purple from one angle and green from another. And we have through a number of different iterations and technology experiments landed on the technology that gets the solar tiles to the point where they're anisotropic. So they blend in with the surrounding non solar tiles in the trim. They look the same from any angle. Exactly. The other thing we did let me if you're in a helicopter. It may look weird if you're in a helicopter. Even from there, it still looks damn good. But the other key aspects was focusing on installation. We really wanted to achieve an installation time that was faster than a new roof plus traditional panels. Well, in fact, we're trying to improve even the installation time for a roof. Yes. So I think with the right tooling and equipment and especially paying close attention to the edge effects of where multiple planes of the roof meet. I think we can and just applying some of the world's best engineering to that, I think we can actually have the solar fast roof installed faster than comp shingle, which is a that's like that's the target. That's been the clear goal? Yes. And we're definitely moving there quickly. We're coming after you comp shingle. The install is on. It's part of our ambition to achieve that. And the key part of installability is looking at how the parts get from the factory to the field to the roof and then how they come together on the roof. So we did a lot of iterations on that. And then the final thing I would mention is, the flashings, the sort of like the edges of the trim, the trimming of the roof, I would say in early versions of the product was almost like custom handicrafts. And what we've done is thought a lot about those problems and come up with beautiful solutions that are very achievable in the field. Yes, totally. This is Kunal. Our installers have been working hard installing this. I've already seen this is much simpler, faster and a lot more intuitive to install. So all those things that you guys have said have led to really fast install times and we are targeting even faster than comp shingle. So yes. Yes, I think I'm confident we'll get there actually. But there's quite a bit of R and D just in the installation process itself. And the difference between having a few simple tools one way or the other, it can dramatically affect the install cost and time. Right. It's also important to have the least amount of bother. It's hard to be in your house while a new roof is being put on. So having that occur within an 8 hour doing like basically an entire roof in 8 hours is quite important because then you can literally come back and go to work, come back in your revenue roof, that's generating electricity. Well, that's our target, sort of 8 hour roof. I think since that time, I'm a roofing company now. Yes, yes. So it's kind of like a roofing company. It's sort of it's a roof. Yes, buildings. Buildings, yes. So yes. Alexander asked, when can we expect Solar Roof B3 to be available for wide residential adoption? Yes, that's now actually. Yes, so you can just literally go to the website and enter your address and place an order right now. We'll be doing installations, ramping up as fast as we possibly can starting in the next few weeks actually. I think actually some are underway right now. And yes, I mean our goal is to obviously get north of 1,000 rigs per week as quickly as possible. So it's hard to predict the early stages of a production ramp because things move as fast as the slowest item. But I think we should be over the next several months well past 1,000 rooms a week. I think that's our goal. Thank you. Robert asked, what is the current solar roof production capacity And how should we think about the production ramp through 2020? How are you feeling about demand? Well, I don't think we're really going to have much of a demand problem, to be honest. I think demand will be far in excess of supply. As always, it's very difficult to predict a supply a production ramp because it almost always looks like an S curve. So it starts with very slow at first because you have a number of constraints and it could be a very trivial part that is limiting the production ramp, but that still will limit the production ramp. It's missing even the simplest item or that's just one little thing is taking longer than expected. That sets the total production rate. So it's essentially impossible to predict with accuracy the fast moving part of the production S curve. But you can predict when the S curve starts to flatten out, That's like much more predictable. And that's why I think you say like probably several months from now, will we be able to do more than 1,000 roofs a week? I think the answer is yes. And then long term, we want to actually do like 10,000 rooms a week and then 20,000 rooms a week. So I think like the addressable market here is something on the order of 100,000,000 houses worldwide. So that's perhaps we want to ramp up as quickly as possible. We're starting off with kind of the textured black glass, and then hopefully, we'll be able to bring to volume production other variants of the solar glass roof every 6 to 9 months, something like that. One of the most challenging is achieving a good sort of earth tone solar glass roof, meaning like a clay tile type of thing. It's just doable, but it's harder to achieve the right look, but that's something maybe is on the order of a year away, something like that. So let's see. Thank you. John asked how will Tesla handle logistics for installations? Do you already have partnerships with installation companies? We don't have any partnerships yet with installation companies. The current plan is to just iron out the broad brushstrokes with our internal installation crews. And so we feel like we're hitting kind of a flat part of the learning curve and then bring in outside roofing companies and installers and say, okay, how can you help us make this better? And so I think we'll see a very rapid improvement in the solar roof and the installation timing. But like I said, Version 3 is really this is like the first version that we think should be ramped up at scale. And this is often true of new technologies like Windows 1 and 2 didn't really work, frankly. Windows 3 was the first big one. So like 3.1 was the main. Yes, I know. 3.1, true. Old school. 3.1. We needed the 0.1 release. The 0.1. The finishing program. 3 minuteus really giant bug. That's true. So if we think of version, hopefully, it would be 3.1 in our case, but we're going to keep improving the roof as we scale. We're hiring a lot of installers. Yeah, we're hiring a lot of installers, yeah. As well as we'll be partnering with installation companies in weeks to come, so lots of action deploying this product. Thank you. From John from say.com, will Buffalo manufacture more than just solar? Any plans to add cell production and manufacture Powerwallpacks there as well, so GigaONE can focus on vehicle sales? Sure. Yes. So GigaTwo is already producing more than just solar. So all of the superchargers are actually manufactured there today, along with the inverter for a power pack and the power electronics for the upcoming mega pack. So yes, there's a real center of excellence of power electronics there, which couples well with solar when you think about it's frequently coupled to the solar. So, yes, and regarding beyond those solar roof and solar and power electronics, we don't have any plans at the moment. It's really primarily the Tesla solar glass roof. That's what our portfolio factory will focus on. But this will be for consumption worldwide. We have thought of in the US because we just need to have a tight feedback. If there are any issues, we can fix them rapidly and then scale, but this is really something we provide worldwide. Thank you. Our first question from the phone lines comes from Kyle Field with CleanTechnician. You may proceed with your question. Hey, it's Kyle from CleanTechnica. Thanks for taking my question. You're welcome. The cost of the solar roof has been a big barrier to date for version 2. Can you speak a little bit about the relative improvement in cost from version 2 to version 3 of the solar glass roof? Yes, you're right. Version 2 was too expensive. It's not like we were making a ton of money. We were just basically trying to not lose money on it. It just wasn't a version that was worth scaling because it was just too expensive. So I think Version 3 is something where we do think we can get below the cost of an average roof plus retrofit solar panels. And feel pretty good about our cost structure there. So it's worth going big time with this version. Okay. We actually have Sorry, I should also say like, one of the things that's obviously delayed to Tesla Solar in general is that we had to focus if we're about 18 to 24 months, we had to focus the entire company on Model 3 production ramp. And so that really kind of stripped resources from solar for a year and a half or thereabouts. And because we had to make Model 3 work or Tesla wouldn't exist. And now that Model 3 is rocking relatively smooth operation, we've been able to redirect resources towards solar and stationary storage. Sure. No, that makes sense. Actually, we have version 2 of the solar roof on our house, and the installation process looks like kind of the Lego version of building the Death Star on the roof. Can you talk a little bit about what you guys have done to streamline that process because that feels like a major barrier for cost and just the speed to installation. Sure. Yes, I can talk about some of the installation improvements. One of the big things that Drew and Ynon both touched upon when you installed V2 was that when 2 roofing planes met or the roof plane had an edge, what we call the edge conditions, we had to like cut the tiles in V2 and that was a very cumbersome process. It took a lot of time, created a lot of wastage on-site, needed extra tools. We've gotten rid of all that in V3 and that's a big major improvement. Just total number of material and parts reduction from V2 to V3 has massively reduced the number of touches our installers have to do on-site and that reduces the time they have to take to install. So there's tons of tiny improvements like that we've made across the product to make the installs really fast. And that's the advantage of Tesla having a vertically integrated chain from we have a team of engineers and installers working day and night at the test home and giving rapid iterative feedback to improve minor installation efficiencies and we'll be doing a lot more of that. Yes, these edge conditions were basically a nightmare on Version 2, and it was quite artisanal. It was like some artisanal glass factory. A lot of like the final production was actually at the house. It's sort of like if you're building airplanes and you finished your production on the runway. It was not great. Yes, we saw some of that. Appreciate the question or the answers. Thanks guys. No problem. Thank you. Our next question comes from Philip Lowe with CNBC. You may proceed with your question. Hi, Elon. Quick question on the installation crews. Can you give us some sense of how many crews you currently have? And what's the staffing per crew? Is it 3 people? Is it 4 people? And are they primarily based in California right now? Give us some kind of a geographic sense of where they are. Well, I'm not sure that's like we're really trying to be product focused here. And I think our current like number of crews and I'm not sure that that's particularly helpful to end customers. Like this is really meant to be, call it, explaining the product to the consumer as opposed to some sort of finance call or something like that. Well, I'm not looking for a finance. I'm just trying to get a sense of, you guys are talking about wanting to ramp up and we're trying to get some sense of what that ability is right now in terms of, hey, we're planning on having 2,000 crews, 3,000 crews. What's that capacity like? So let's put it this way. We installed Tesla Solar in 25 states in the country. We'll be definitely installing solar roof in all those states because we already have energy warehouses and we'll be installing solar roof there as well. And as Elon said, we'll be expanding across the country with our own crews as well as 3rd party crews. So that's all we can say right now. But we are in 25 states installing solar and Solar Roof will be there in all those states as well. Yes, that was the next question. Thank you. Our next question comes from Sean O'Kane with The Verge. You may proceed with your question. Hi, guys. It sounds like there is a little bit, while you simplified some of the stuff that makes it a little easier to install the new version, There's maybe a little bit more going on in the tiles themselves. Is there can you tell me what sort of process you went through to make sure you don't run into the same sort of production issues that you had with previous versions? Yes. I think as The issue with prior versions was not so much production, it was the installation was problematic. So we could have ramped up production of Version 2, but it would have run into a fundamental obstacle with the difficulty of installation. But if you're wondering like will we run into difficulty ramping Version 3? Absolutely. We have a massive amount of difficulty and we'll go through a ton of pain. And there will be like setbacks. I want to be clear. We have applied some preventative measures, however, by reducing the number of parts and subassemblies by half, we've dramatically reduced the number of processing steps and manufacturing steps, which simplifies any manufacturing ramp. And I will say that the early production that we are in right now has gone smoothly. And now as we ramp on the S curve, we will find things that we have to overcome, but the road production looks good so far. Yes. It's always like quite sort of jumpy and thorny on a short term basis. But I think a year from now, it will be pretty smooth sailing. And Alex, if you also often just get too focused on the short term. I think if you want to understand like where our products are going to go a long time, it's just what do you want the world to look like? And it's like, do you want a future where you look around at the houses in the neighborhood and they're generating electricity and they're beautiful and they're long lasting? Of course. So that's the future we need to make happen. Thank you. Our next question comes from Kim Javehary from the YouTube channel, Like Tesla. You may proceed with your question. Hi. I was curious if you could elaborate a little more on what the warranty is like for the tiles? And if you're using a 3rd party roofer, how that will affect the warranty? Yes, sure. Yes, this version we are offering a 25 year warranty, which is weatherization, power and we have 130 miles per hour wind resistance, hail resistance, so 25 years is the warranty right now. Yeah, these are things that are fairly standard in the roofing industry. And then it's more durability. Yes. We definitely don't have all the answers right now. And it's kind of like a to be frank, it's kind of like a weird and sort of odd product. You don't really hear about it. It's like why would anyone make a solar roof? How strange. But it's just the thing that should be. And so we're going to make it. It's going to be real confusing because there's like no actual product like this. That's the nature of 3rd party. My one comment on, like, if we have 3rd party installers, how will the warranty work? I mean that is how roofing works today, right? There's like a manufacturer of the tile or the asphalt shingle square. And if the certified installer or roofer has been trained and has been audited to install correctly, then eventually the warranty is extended to that installer. But I wouldn't expect us to do anything different with this product. Okay. That makes sense. There's quite a big market for roofs actually. I mean, on the order of 100,000,000 roofs in North America and they get replaced about every 25 years. So it's about like maybe 4,000,000 roofs a year. And 5,000,000 new homes a year. I mean, sorry, 1,000,000 new homes. Yes, 1,000,000 new homes a year. So call it like on the order of 5,000,000 in North America, it was just 5% of the world's population or thereabouts. So it's quite a big market. This is why I think we're going to be production and sort of installation constrained. It's going to I mean, on a percentage basis, it will grow like CALP on steroids, but it's just business like so many things that have to be sorted out. But I think it's going to be a very exciting product and it's like this kind of thing that I think people will want their roof to be this way. Thank you. Our next question comes from Roberto Baldwin with Engadget. You may proceed with your question. Hi. I'm curious on the energy efficiency of the tiles versus the panels. So if you have, let's say, a 200 square foot area of the tiles and 200 square foot area of the panels, like what is the difference in the sunlight to energy conversion? Sure. There's a slight impact, Virginia. Well, the target is equivalent to or better. There are some benefits to the Solar Roof in that we have a smaller like integer size. So when you think about The granularity. We can basically put a question of like what's the cell efficiency. If you're purely looking at on a cell efficiency basis, you're also going to take some knockdown in order to have better aesthetics. That knockdown, we want to try to keep that knockdown to something on the order of maybe 10% -ish, hopefully less, Bruce likes saying less. But from like a best case, like ideally, it's like a single digit effect on the energy efficiency on a cell level, but you can put far more cells on the roof and do so without affecting the aesthetics. In fact, you can't even tell which ones are active and which ones are which paddles are active, which ones are not active. So the total power capability of the roof is significantly greater with the solar glass roof. Yes. And if you haven't ever had solar, I mean, just drive around your neighborhood and look at homes and you can see the quantization problem of the panel. And you've got skylights, you've got stovepipes. Roofs are messy and complicated. Yeah. And by the way, generally look terrible. You don't really don't pay attention. You don't really don't pay attention to roofs, you're like, wow, we had a lot of trouble finding good looking roofs, because we like we want our roof to be the best looking roof and it's not easy finding pictures of good roofs. Let me tell you, it's very difficult and this also is a giant forecast. So for all of those reasons, we have equivalent or better and better in almost all cases, power generation. Thank you. Yes, absolutely. From a consumer standpoint, it will be a better outcome. If the right move, if you're going to get a move and the solar panel, This is a solar glass integrated. So it seems to be a better outcome, we believe, in controlling economic power generation, aesthetics and verticals. Yes. And I just want to add one more comment. We're not married to a particular solar cell technology in our aesthetic solution. So as solar cells get better, the solar roof will also get better. And that was a very important target for the team to make sure that we cannot take ourselves on a legacy solar technology while the industry is important. Solar efficiency is Star and flow. Star and flow. Yeah. Yes. Thank you. Thank you. Our next question comes from Daryl Ethington with TechCrunch. You may proceed with your question. Hi. Thanks for taking the call. I was just wondering, do you see any other potential formats or incarnations of this technology that might enter the market in different ways or be offered to consumers in different ways? Yes. We're really just scaling this one variant for now, which is the kind of the textured black glass look. But then over time, we want to have something that like this has sort of an earth tone, sort of more like a clay tile. We want to have one that looks like a French slate. And because not all houses should have like black textured glass, especially if it's sort of more of a Mediterranean architecture, it's going to look out of place. So making that work is next on the agenda from a product development standpoint in terms of a significant improvement or significant, I should say, variant, not improvement. It's like we're getting earth tone version that looks right is not the next major challenge for the engineering team. But I think we'll get some time probably next year. Thank you. Our next question comes from Hope King with Cheddar. Hi, thanks so much for taking my question. So from a consumer standpoint, especially we've seen happen in California and EPG and you said, you could take another 10 years before some of these outages are ratcheted down significantly. And Elon, you've tweeting recently to get folks to order Tesla Solar and Powerwall. Are you already seeing orders grow, demand grow as a result of the outages? I think we are seeing some demand growth as a result of that, yes. So that's the reason that when you turn the lights off on people, they don't want that to keep happening for the next 10 years. So and there's no obvious solution. And so that when you're just sort of sitting there in the dark and all of your devices are battery powered, when you lose your phone connection, it's like a security risk. You can't even call 911, like if there's a problem. So we're definitely seeing people interested in solar plus Powerwall batteries, so they can have no blackouts and have renewable energy. And yes, so that's definitely what we've seen demand increase as a result, and I think we'll continue to see that in the future. And just how important is this also to developing the overall Tesla energy ecosystem that you guys want to build out? I think this is very important. The fundamental goal of Tesla is to accelerate the advent of sustainable energy. So we have to therefore generate the energy in a sustainable way, which really means solar primarily in our view. So that's both solar at the residential level, at the commercial level and at the utility level. And then since the sun obviously doesn't shine at night and you have like cloudy weather at times, you have to have storage to have continuous power. And so there's really 3 parts to the sustainable energy future. From a Tesla standpoint, solar power generation, storing that solar with stationary storage of the Powerwall Powerpack and Megapack, and then consuming that energy in a sustainable way, which means electric transport. And I think the fundamental good of Tesla, in my view, is to what degree did we accelerate the advent of a sustainable energy future. Thank you. Our next question comes from Carissa Bell with Mashable. You may proceed with your question. Hey, thanks for taking my questions. You've talked a lot about kind of how you're improving the installation process, making that more efficient, a lot faster. I'm wondering if you can kind of talk a little bit more about sort of what you're anticipating the wait time will be like for people who make a reservation or pre order for this and kind of how long you're anticipating that they'll have to wait between that and when they'll actually be able to get it installed? Well, it depends on how many people sign up for the solar glass roof. If it's a lot of people, then we'll just sort of queue things accordingly. It might be a few months that you'll have to wait or several months. I think the sooner people sign up, the less time they'll have to wait. And we will grow this exponentially. So it might be like doubling every month or something like that. So we want to get to the point next year where there's a very short wait time. If it's possible to do so, we'll do our best to achieve that. Sorry, go ahead. Do you think that the installation process will help speed things up overall that this should be like a much more efficient end to end from reservation to installation than it was with version 1 or 2? Yes. This is really intended to be easy to install. And there will continue to be improvements that we make to Version 3 that actually are mostly related to the edge conditions, like potentially the corners of the roofs where the planes of the roof meet and at the edge of the roof where the rain gutters are and that kind of thing. So those are very rapid evolution of those edge condition details and that's, I think, what is the key to growing this rapidly. And like I said, getting our install time for a typical roof under 8 hours, which means it's better to basically have a fairly big crew attack the roof all at once, like surgical strike and just get it done quickly versus a small number of people take a long time to do the roof because that minimizes having it be done fast just minimizes the inconvenience to the homeowner. And there's less setup teardown time every day. Yeah. Exactly. Fewer truck rolls internally. And we'll be installing starting next month in California, and we already have customers lined up. Yes. One other comment on the wait time. We are actually employing some now, but yes. We're installing now. The wait time, one thing to consider is weather. So in places where it's like very covered in snow, like obviously there's some wait time there, but correct. Very rainy up on But I mean, I guess the intention here is this is a roof. We want to install it faster than traditional roofs are installed. And so the benchmark is that calling somebody up saying you want a new roof and getting the roof installed like as soon as the weather is good, that's exactly how our products will work. Yeah, it seems like seamless and easy and not have to wait a long time. All right, cool. I think that's probably our last question. Thanks everyone for tuning in. We look forward to installing ultimately millions of rigs. Thanks. Thank you. Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.