Today’s satellites are cheaper, but hardly cheap. Brad King, the CEO of a propulsion supplier Orbion Space Technology, joins us to explain why after a writing a column for our Oct. 13 print issue.
Subscribe Now
Don't miss a single episode of the award-winning Check 6. Follow us in Apple Podcasts, Spotify or wherever you listen to podcasts.
Discover all of our podcasts at aviationweek.com/podcasts
AI-Generated Transcript
Joe Anselmo: Welcome to this week's Check 6 podcast. I'm Joe Anselmo, Aviation Week's editorial director and editor-in-chief of Aviation Week and Space Technology Magazine. The cost of access to space has declined dramatically in recent years, paving the way for an orbital economy, but while satellites are cheaper, they're not cheap. Most are still priced in the millions of dollars, which begs the question, why are satellites still so expensive? Brad King, the CEO of Orbion Space Technology, a Michigan-based satellite propulsion supplier, joins us to tackle that question. Also with us from Seattle is Garrett Reim, Aviation Week space and emerging technologies editor. Welcome, Brad. You've written a viewpoint for the next edition of Aviation Week, and you make this point that a satellite that's about the size of a refrigerator in a college dorm room still costs 10 times as much as a handbuilt Ferrari automobile. Why is that?
Brad King: That's a good question, Joe. The mystery there in satellite pricing is what motivated me to write that article. Over the past handful of years, the question comes up all the time from laypeople and other engineers and they say, we hear satellite costs are coming way down. Not, yeah, they're coming way down. Well, what do they cost? They say, well, they're down to maybe $5 million or $10 million a vehicle. And then the eyes pop up and they say, well, why? Well, we don't make many of them. It's just a couple thousand a year. They're handbuilt, they're not on assembly lines. And then often the response is, well, so are supercars and I can get a Ferrari for $400,000. Why $5 million or $10 million for a satellite? And that was kind of the introspection that started that article. And we tried to pick apart what we've learned in the industry from the past 10 years and identify the key players in that. And as we'll talk about here, a lot of it is reliability and reliability within a multi-tiered supply chain.
Joe Anselmo: So what areas can we reduce costs further to get those prices down?
Brad King: The easy way to do it conceptually is go completely vertical. As you know with any kind of a supply chain, you've got raw materials on the bottom, you've got piece parts, you've got components, you've got subsystems all the way up, and everybody in that supply chain wants to make a margin. So if you can go completely vertical, you can save all those margins. And that's often thought of as sort of the SpaceX model. SpaceX loves going vertical and people almost tongue-in-cheek say, Elon's going to start buying tungsten and aluminum mines so he can control the supply chain all the way top to bottom. That's a good way to bring the recurring cost per unit down, but the capital required to put that vertical supply chain together is enormous. So that's not always feasible to everybody, but some degree of consolidation in the supply chain would definitely help that. And I don't know what the right degree of consolidation is, and that has two levers that it has a lever on the margins, as I mentioned, and a lever on reliability testing that we talked about in the article.
Joe Anselmo: And another point you had brought up is risk. Part of the reason that satellites cost so much is that once you launch it into space, you can't get it back to fine-tune it. So there's an awful lot of testing that goes into these satellites, isn't there?
Brad King: And again, thinking of the SpaceX model or think of a university program or a new startup, let's say if I'm just building for myself, if my company, we've got a satellite we want to put up there, we want to reduce the cost and we want to prove some kind of concept by getting it to turn on and do something, we're willing to take quite a few risks on that. We're willing to waive a bunch of reliability testing and get that thing up because if it fails, well, we knew the risks going in and we were well aware of them and maybe our experiment didn't work. The problem is that those of us in the supply chain, we're not building for ourselves. I can't hand over a product to my customer and say it's low cost and it's probably going to work, so go for it. The risk when you're building for yourself, your tolerance for that is much higher than the risk when you're building for a paying customer.
Joe Anselmo: Garrett, welcome. You recently wrote a cover story for us on chasing Starlink, how everybody is chasing Starlink, but there was a point you made in there that I found fascinating. You note that not long ago, electronically steered antennas cost hundreds of thousands of dollars and Starlink selling them now for as little as $349. Pretty amazing stuff going on.
Garrett Reim: Yeah, that's certainly an example of vertical integration. SpaceX makes their own antennas. They do it at a massive scale in Texas, they do the same thing with their satellites. To touch back to what Brad was talking about, different ways of quality assurance. SpaceX does a batch testing program, so they statistically sample their production and they don't individually test each satellite, and so they accept some statistical amount of failure and they tolerate that, and that helps them bring their cost down. They just don't have to make sure each individual satellite is rigorously accepted.
Brad King: And then Garrett, that's a great example of building for, they can take advantage of that statistical sampling and sample test, maybe one in every 1,000 and be reasonably assured that the other 999 are OK. That doesn't work when you're selling to another customer. If I tell a customer, Hey, I test one out of 1,000 and that one was good, so you're probably all right, the customer's going to come back to me and say, no, no, no. I want you to test every serial number you're giving me and I want you to stand behind that, which I'm happy to do, and then I need to pass that cost of testing onto my customer.
Garrett Reim: Yeah. I wonder, is there some sort of world in which you can be more explicit with your customers about the risk that they would take with a lower cost batch testing system and maybe there's some sort of insurance that they could buy alongside that where they could take advantage of that quantities of scale, but then also take some risk out too.
Brad King: We talked through some of those. There's a couple challenges there. First of all, most of our customers feel that they're very risk tolerant. Hey, tell me how you can reduce costs because we can tolerate risks when push comes to shove, they can't. They can tolerate taking risks, but they can't tolerate the failure if the risk turns out bad, the monetary hit of this thing failing on orbit. So they love to roll the dice up front and save 25%, but it could crush them if it fails. So when you look at that kind of calculus, they then start to get pushed into a corner and go, no, no, no, you got to prove to me that this thing is going to work. And then insurance or some kind of warranty is sort of an unknown liability. So if I'm going to test only one out of every 1,000 and pass it on to my customer for a low cost and say, tell you what, if some of them fail, I got you covered. Well, that's a cost I have to build in as well because it's an insurance model, and I know sooner or later that cost is going to catch up with me and I haven't planned for it. So it's not a great way to reduce the cost in the long term.
Joe Anselmo: Brad, you also talked about the multiplier effect. Basically every piece of the supply chain needs to get a little bit of a cut and that just rolls and makes it more expensive, right?
Brad King: Yeah, I did some real simple math and it's caveman math, but it's meaningful. So if you consider a five-tier supply chain from piece parts up to the top and the bottom of that supply chain is just the cost of the raw materials. Every supplier needs to stay financially viable, so they're going to want to put some margin on it. So maybe they put a 50% margin on the raw materials. And then if that supplier is required to do reliability testing, reliability testing for space is expensive and it's bespoke. That can add another 50% on the raw material cost just to do that testing on every serial number. So you got a 1.5 times a 1.5 multiplier on every tier. You bring that up five tiers, and you got a 25x cost multiplier on the raw. And that multiplier is there to ensure everybody in the supply chain is responsible and accountable for their reliability, and they all stay financially viable to answer the phone the next year. So that's kind of the unspoken consequence of a multi-tier supply chain with responsibility and margin.
Garrett Reim: Yeah. I wonder too, if this will just come in time, if the industry is just immature, and that over time as the supply chain becomes more diversified and sophisticated and people are more just aware of different risks and different products that you may see costs come down, you see emerging suppliers like Redwire or Rocket Lab where people now they don't have to build everything. They can go to a company like that to buy solar panels for instance. And maybe over time those guys get enough scale and that really brings down cost. What do you think about that? How is the supply chain going to evolve and are we stuck in this situation or does the supply chain eventually become more sophisticated and then that brings costs down?
Brad King: I think there's two ways. The first is to start to compress the supply chain and have more of these nearly vertical providers. So every layer, every tier you take out of the supply chain, you win back cost by 2x. So instead of a raw material, a piece part, a subsystem, if somebody can blend three of those tiers, you got a really good savings right there without hurting anybody. So some of that's going to happen naturally, as you mentioned with Redwire and other folks consolidating the supply chain. The second is the reliability testing. If there can be some better standards for when and how that reliability testing happens, maybe it doesn't have to be at every level. In fact, testing it at every level, we found surprisingly can threaten reliability. So for instance, our piece part, we sit in about the middle of the supply chain and we have to do tests like you fly.
We've got to prove when we hand the serial number over to the customer that it's going to survive launch. So we take this thing and before we deliver it, we vibe it, we thermal cycle it, we performance test it, and we hand it over with an end-item data package that proves that this unit met those requirements. What we were not really aware of five years ago was now our customer was going to take this and bolt it onto their satellite bus, do the exact same test, thermal cycling and vibe it again, then they're going to put the payload on and the satellite customer is going to do the exact same thing, thermal cycling and vibe it again. So all of a sudden, by the time our component goes into orbit, it's been vibed three or four times and we weren't really planning on that. So there is a risk of damaging it by over-testing in this hunt for reliability. And I joke that often the hardest thing about getting hardware into space is surviving the supply chain testing. So testing, there are some push and pulls that happen in that sequence.
Joe Anselmo: So Brad, you're not seeing a problem with business. I mean, business is booming. There's lots of demand for small satellites at today's cost, isn't there?
Brad King: There is. And one of the viewpoints I took in the article is that maybe the industry just has to come to grips with the fact that if you are not fully vertical, $5 million-ish for a satellite with nearly guaranteed reliability is the right price. And if that's the case, then we adjust business models accordingly. There are still some companies out there chasing the possibly mythical $500,000 satellite. They say, this is all nonsense. I know you can build that for half a million dollars and I'm going to build a business model that if I can buy a thousand satellites for a half million dollars, we're going to be hugely successful. Well, if that's never going to happen, you need to adjust your business model. If $5 million is the price you're going to have to pay, then stop expecting the mythical half-million-dollar satellite.
Joe Anselmo: OK. And Garrett, you recently wrote a series on manufacturing and aerospace. You visited lots of these small suppliers, space suppliers. There is progress being made, isn't there?
Garrett Reim: Yeah, there is a lot of progress and people are trying to tackle this problem in a lot of different ways. There's the vertical integration approach. There's suppliers who are trying to reach scale and sell, trying to integrate more subcomponents, so you get kind of more of a modular solution. So you're not just buying one thing from somebody, you get a satellite bus plus solar panels, plus other subcomponents all kind of together. And I think this kind of goes to what Brad was saying about a nearly vertical supply chain where you're consolidating a bit. My sense is that there's so many bets and so many players at this point that you're going to see some consolidation. There are a lot of companies that are trying to be the satellite bus manufacturer, and can we have, I can't remember how many, I haven't done a count, but there's dozens of people who do satellite buses. Does that make sense long-term? Especially for something that's supposed to somewhat be standardized. So yeah, Brad, I don't know if you want to jump in on that, but it seems like there's a lot of people are trying to tackle this problem in a lot of different ways, and I think eventually we're going to start seeing some preferred solutions.
Brad King: We see that as well, and I don't know what the winning path is going to be, but we often question when a new satellite bus manufacturer comes out and their business line is we're going to build low-cost satellite buses, and then their operation model is, and we're going to do this by buying all the components from others and integrating them together on a bus. And then the head-scratcher is you're buying the same components everybody else is buying. Where's the magic? Where is the time travel that allows you to buy these same components and cut the cost by 5x without going vertical and learning how to build your own reaction wheels, your own propulsion systems, your own solar panels, your own batteries. If you can do that vertically, then I think you have a shot at a very reduced-cost bus, but that could be a couple billion dollars in capex to bring that vertical capability.
Joe Anselmo: If someone does accomplish that, is that going to knock a lot of the traditional suppliers out of business because it can be done so much cheaper?
Brad King: Well then it's just business 101. Once they go vertical, they've got a giant capital outlay to go vertical. They can sell these things at a lower cost, but they've got to make up that capital outlay and amortize that over the next few years. And if the revenue model and the profit doesn't close, you're just left with a really big expensive failure.
Joe Anselmo: Garrett, you're nodding your head.
Garrett Reim: Yeah, no, I mean there's a ton of venture capital pouring into this industry. There's a ton of Defense Department money pouring into the industry. At some point, there's going to be a reckoning. That's the benefits and the cost of free market economy. There's going to be winners and there's going to be losers. We're kind of in a bubble. I think in some places we don't need as many suppliers as we have, and so at some point, but we don't really know who's doing stuff that's not so valuable. That's being just propped up by short-term contracts or venture capital. Eventually we'll start figuring that out, but it's going to take a while to sort it all out.
Joe Anselmo: OK, Brad, last question. I talked about Starlink's electronically steered antennas. Are there any other opportunities that you see where we could radically cut costs?
Brad King: Oh, geez. I wish I knew as a supplier, it's changing a lot in the last couple of years, but for the first few years, we would have customers come to us and ask us to quote them prices, and their immediate reaction was, oh my God, they must be hiding an enormous margin. Let's just bargain harder and get this price down. And that philosophy was kind of across the space industry for a while, and now I think people are realizing that suppliers aren't hiding enormous margins. In fact, there's been a lot of margin compression as everybody is struggling to get market share. If there were some magic way I could reduce the cost of my product by 50%, I would absolutely do that. So in some of these instances, I don't know where the ESA cost reduction came from, but at least in our technology niche, we don't see any magic there.
Joe Anselmo: OK. Well, we're just about out of time, but I wanted to thank you, Brad, for taking the time to join us and also for writing your viewpoint. That is a wrap for this episode of Check 6. A special thanks to our podcast editor in London, Guy Ferneyhough. If you haven't already, be sure to follow Check 6 wherever you get podcasts, so you never miss an episode. You can also read Brad's column on aviationweek.com and in the Oct. 13 print edition of Aviation Week Magazine. If you found today's discussion helpful, please consider leaving us a star rating or review. Better yet, share this episode with a friend or colleague. Thank you for your time and have a great week.
Related Content
