Goldman Sachs Industrials & Materials Conference 2025 - Adam Spice Interview

Thanks for having me.

Well, look, I think it kind of goes back to if you're going to be in this market, we think there's been way too many examples of companies that have kind of brought faulty products to market that weren't ready for prime time. And the appetite for mission failure is zero, right, with our customers. I mean, we know what happens. It's the nature of the launch business to a certain extent, but you don't ever willingly go in knowing you want to take more risk than is prudent. And I think Pete is, he's a hardcore engineer at his core. He understands the product at a level that most other CEOs, I think, probably don't have that familiarity because he designed, he built with his own hands Electron, which has now launched, I think, 77 times. And with each one of those launches, you learn so much and you understand kind of where the risk is in the vehicle. And he just, in his process of developing a new rocket, he's leveraging everything we learned in Electron to make it as successful as it was. And he's just not willing to cut any corners to save schedule or to a certain extent cost. He's like, look, people won't remember a three-month push to a schedule, but they will remember a big smoking hole in the ground.

So, he's absolutely not willing to take that risk.

Yeah, so we talked in quite an amount of detail on our conference call about the timing. Right now the goal is to get a vehicle to the pad in Q1 and then launch as quickly thereafter as we can. So if you think about what it takes to bring a program like a medium lift launch vehicle to market, there's a tremendous amount of coordination amongst the various disciplines, right? So you've got engines and avionics and tanks and pad infrastructure and everything else that has to come together. And the one thing that's created a little bit more of a challenge for us with Neutron than we experienced on Electron is just the scale. If you look at some of the pictures and maybe if some of you folks have been on some of the forums lately, I think in the last couple days, there's pictures of the Hungry Hippo fairing making its way through New Zealand. Ultimately, that fairing is going to be manufactured in the US in our facility just outside of Baltimore. But the initial prototyping was done in New Zealand because that's where a lot of our carbon composite expertise reside. I mean, these things are huge. I was looking at another one of the pictures this morning and you've got this massive transport truck pulling this huge fairing that's five-and-a-half meter diameter fairing and quite long and everything else. It's just moving parts like that and getting them through their qual is taking so much longer. When we were doing Electron, if you were going to qual something or want to do a tweak, you literally could pick up an engine with your hands, right, and carry it across the factory. Now, an engine weighs more than a ton, right? You need the cranes and all the different things. And when you're moving these huge carbon structures, you want to be careful that you're not putting stress on a joint that could take a long time to fix if you break something. So, we're being very ginger with all this stuff to make sure because we understand how important schedule is. But it's until you're faced with the reality of staring into this huge structure that you've got to move and go, 'Yeah, that fixture that we were thinking was going to do the job probably needs to be augmented a bit here and there and it's just taking a little bit longer.' So, the plan is to get everything to the pad in Q1. What does that mean? That means you have the rocket that gets stacked and integrated on the pad, which is a little unique. Other times it'll be done in integration facilities. We're actually stacking Neutron on the pad. So having that ready to go and then now you're in the position where you're kind of running your final tests and you're getting your final license from the FAA and NASA. So I think we've got a pretty well-defined path to get where we need to go. But rocket to the pad in Q1, launch as quickly thereafter, the final checkouts and allow is really the goal.

So the program as we exit 2025, we will have spent about $360 million and that's across capex and R&D. And so if you think of what that has accomplished, it's got a brand new pad at Wallops. It's gotten new integration facilities at Wallops. It's gotten a structures complex outside of Baltimore. It's got static test — it's actually been a very efficient project from that perspective. And so if you kind of target a first half of 2026 launch, I think there's probably about another $40 to $50 million of spend again across R&D and capex to get us to that first launch. So that'll put the program at a total of about $400 million and it will roughly have taken about five years. The original estimate was $250 to $300 million over about four years. So we experienced about a one-year total push to the program and spending is, you know, call it at the high end about $150, but really kind of $100 million above what we originally anticipated. But to put that into context, if you look at Blue Origin had a successful flight of New Glenn a couple weeks ago. Very impressive actually, pretty encouraging for folks that are also looking to do the reusable kind of ocean-landed architecture. You know, this is anecdotal because Blue is a private company, but kind of on the order of magnitude of, I think it's 14 years and probably north of $10 billion of investment. And then if you look at what it took to bring Vulcan to the pad, right, that was also measured in probably almost a decade and many billions of dollars, maybe not as much as New Glenn, but that's the kind of neighborhood that usually these programs occupy and we're talking about doing things in half or less the amount of time and for orders of magnitude less in total spending. So, we're trying to do some very aggressive things and I think we're always trying to push the envelope and Neutron is no different.

You know, I think we're somewhat fortunate in the fact that this is a new and expanding market, but we've got a pretty clear comparable and peer that we can point to with an established run rate in this very segment. So, if you look at Falcon 9, that's really the only other medium lift launch vehicle. Last year, I believe they launched about 134 times. This year, they were talking about launching over 190 times.

Close to 200. But about 60% of those volumes are estimated to be captive to Starlink. So if you kind of take that and you say, okay, roughly you think of 90-ish launches of merchant volume for a Falcon 9, it's been growing pretty dramatically year over year. So look, we think that there's a lot of market opportunity for Neutron. The easiest one to identify is the government side with NSSL launches which we're now getting onboarded into that program. That program was funded to the tune for the lane that we're entering into to about $5.4 billion of earmarked spend between now and the end of 2029. So that gives you one sense of kind of the customer bucket of dollars. In order for us to start tapping that, we have to have a successful test launch of Neutron. Once we have that, then they can start dropping task orders down to the lane for us to bid on. So clearly there's a large volume of launch in that lane that's expected from the government customer, but again, we have to unlock that with a successful test launch. On the commercial side, there's also a lot of demand. Just to again provide a little bit of context, the first wave of deployments of Amazon Project Kuiper, which is now called Project LEO, which we're not contracted for, just to be very clear, but to look at it as an opportunity there, that's I think it was about 90 launches across three providers over the next several years. And so then once that constellation needs to be kind of reconstituted because lower Earth orbit satellites don't stay up in orbit as long as the prior generations of LEOs and so forth. I mean, reconstituting a constellation like that will take many dozens of launches per year. So, you've got a very tangible government opportunity which is just starting to ramp because a lot of these higher volume proliferated architectures like SDA, they're just now starting to be launched and that's going to be a clear kind of present driver of launch demand for the foreseeable future. Then you've got the commercial side with things like Project LEO. You've got things like Telesat Lightspeed and IRIS² and so I think the demand is clearly there and what the market needs is more than one player and that's all it really has to choose from today.

So the feedback that we're getting from customers is it's about two years, right? Say when you want to basically — I'm ready to contract a launch on a Falcon 9, it takes about two years is our understanding. I want to come back real quickly to the sizing of the number of launches available for Neutron.

Clearly there's an existing market today for Falcon 9 for merchant launches which is approaching 100, probably going to be very easily above that in 2026 and beyond. Our target to get to our target margin model for Neutron, given our pricing assumptions and cost for refurb and so forth, it only takes about 10 launches per year for us to actually get to our target margin model for this business. So obviously we think that the opportunity is much, much, much larger than that. But it doesn't take a disproportionate amount of that existing volume today, without assuming the growth potential, to kind of hit our marks.

So I think most people understand that the launch business is a very high fixed cost business. So whether you're launching once or 20 times, you've got a certain amount of infrastructure that needs to be maintained. For us, for Electron, it's about $40 million a year for Electron that maintains our launch facilities. It's all our production overheads and so forth, it excludes the variable pieces of that like the labor to build the rockets. But if you just think about your fixed nut to kind of stay in that market. So if you're launching once a year, you've got a horribly unprofitable business. We've been pretty consistent in saying that if we get to 24 launches a year or two per month for Electron, that's where we hit the sweet spot of our overhead absorption and margins on a non-GAAP basis in the 45 to 50 point range.

Gross margins. So given the guidance that we provided a few weeks ago, which embedded roughly seven launches in Q4, we're kind of there, right? So we don't need any material growth to get us to where we need to be. We're knocking on that door as we speak. And again it kind of just reaffirmed for us — cadence is super important, right? You need to get that overhead absorption rolling. ASP fortunately in our case has also been pretty helpful. It's been up and to the right the last few years and we see that continuing that direction given just the lack of any real credible competition in certain segments of the market. So we think the demand is there. We now dialed in the production and scaling of Electron right now. And even at that 24 launches per year, which gets us to your target margin, we can build twice that. We can basically build a rocket per week with our existing footprint for Electron. So you can only imagine there's kind of margin gravy to come at anything really above that level.

Yeah, I mentioned the fact that part of the challenge of Neutron is just the scale with the vehicle. So, the scale of production is greater, the launch pads are bigger, everything is bigger. So, instead of $40 million per year of kind of standing cost, it's really more like about $80 million of standing cost. So if you think about the margin path for Electron has been pretty predictable. It's always been about growing or maintaining your ASP and getting your cadence up to absorb the overheads. With Neutron being a reusable vehicle, it's less about that and it's much more about how quickly can we get to reusability on Neutron. And that really will have a much more accelerated path to margin expansion than we experienced on Electron where you have to build one for every mission. The power of the fleet model for Neutron is again that's already been established. We've seen what that can do to SpaceX, right? As far as being able to launch, you can launch a lot more if you build less. It's a lot easier to refurbish a rocket than to build a rocket, right? So that's really the whole premise underpinning the margin expansion because the first launch for Neutron that we talked about earlier, that's just an R&D launch. There's no revenue associated with it. A lot of that R&D is already kind of in the rearview mirror. And that first rocket, the goal there is to get it to space successfully, re-entering the atmosphere and do a propulsive soft landing in the ocean. We're not trying to catch that one or land it on a barge. If everything goes really well, the goal would be then for the second tail or second rocket to basically land on a barge. And that rocket will probably not also be used. We'll use that for kind of postmorteming and kind of really dialing in the block upgrades to the long-term architecture of the vehicle if there's any major changes or minor changes. And so it's really the third tail that we have the opportunity to land it on the barge and then put that into reuse. So you can kind of think about it — it's the third tail onward that represents the real margin expansion for that business. That's where you can start to see that trajectory, but the first couple of launches are going to be negative gross margin propositions for sure.