Rocket Lab Corporation (RKLB) Earnings Call Transcript
February 26, 2026
Earnings Call Speaker Segments
Thanks. Hi everyone, and thanks for joining us on this call, where we'll go over Rocket Lab's results for the fourth quarter and the complete 2025 fiscal year, along with business highlights and other updates. Before we get started, please keep in mind that what we say may include forward-looking statements about how the company will perform going forward, and these are meant to fall under the safe harbor protection from liability set up by the Private Securities Litigation Reform Act. Statements of that kind aren't promises about future results, and the factors that might affect our results are called out in today's press release, while others appear in our filings with the Securities and Exchange Commission. Those statements rest on information the company had as of today's date and could change because of future developments. Unless the law requires it, the company takes on no duty to update them. Today's remarks and press release also include non-GAAP financial measures as defined in Regulation G enacted by the SEC. That release and our supplemental materials contain reconciliations of these historical non-GAAP financial measures to the comparable financial measures calculated under GAAP; in addition, this call is being webcast alongside a supporting presentation, and a replay plus a copy of the presentation will be posted on our website. Speaking today are Rocket Lab's Founder and Chief Executive, Sir Peter Beck, and Chief Financial Officer, Adam Spice. They'll cover key business highlights, among them updates on our launch and Space Systems programs; we'll go through financial highlights and our outlook, then wrap up with questions. With that said, I'll hand the call over to Sir Peter.
Thank you so much, Morgan. To kick things off today, I'm going to borrow a bit of Adam's spotlight and walk through several of the financial standout items right at the top. During 2025 we set a fresh all-time high for yearly revenue, landing at $602 million — that works out to 38% growth compared with 2024 on a year-over-year basis. We likewise posted a record-setting quarter in Q4, with revenue reaching $180 million, an increase of 36% over Q4 of the prior year. When Q4 wrapped up, our backlog stood at an all-time high of $1.85 billion, marking a 73% rise versus the equivalent point in 2024. To wrap up, Q4 brought us our best-ever gross margins — 38% on a GAAP basis and 44% non-GAAP. To borrow your typical launch-day phrasing, every board shows green, and the result is exceptional. One factor explains all of it: the Rocket Lab team's unrelenting execution across our launch and Space Systems programs. A few standout points from that execution follow. I'll keep this brief for now, because we'll cover things in more depth on the slides in just a moment. Ultimately, though, we set records for both the number of electron missions launched and signed, moved fast to lead the way in hypersonics testing, and simultaneously cleared several important qualification and development milestones for Neutron. On the Space Systems front, we secured Rocket Lab's largest-ever contract, got NASA's ESCAPADE mission to Mars delivered successfully, and experienced record expansion in every one of our Space Systems component businesses. Regarding acquisitions, Geost joined us in 2025, officially marking our entry into payloads, and then in Q1 2026 we acquired Optical Support, Inc., which bolsters our optical systems offerings even further. We also broadened our machining and manufacturing capacity by acquiring Precision Components Limited, a deal that actually closed today and will in the end help keep scaling up that components manufacturer for both Launch and space systems. We'll go deeper into those points on upcoming slides. For now, here are a few brief updates on Electron and HASTE. Globally, Rocket Lab still holds the top spot in small launch, since it is the sole rocket providing dependable, frequent launch windows for SmallSat. During 2025, we flew 21 missions combining Electron and HASTE, setting a fresh company record. In Q4 alone we flew 7 missions, the largest count we've achieved in any one quarter so far. At the same time, no new small launch vehicle from the U.S. or Europe managed a successful orbital launch in 2025. You know, it's pretty obvious that when small cell operators require a ride to orbit that's dedicated, Rocket Lab is who they turn to. We're proud to hold that title, and we expect to extend that record even further over the course of this year. The U.S. government hasn't hidden the fact that testing hypersonics more quickly and more often is a pressing need and something the nation treats as a priority. When it comes to actually delivering this capability today—not somewhere years down the road—Rocket Lab stands alone as the only provider with real credibility. We successfully flew three HASTE missions in 2025, and the next one is sitting on a pad in Virginia as we speak, just days away from launching. That sort of tempo and dependability puts us in a strong spot for efforts such as Golden Dome; since additional HASTE missions are scheduled this year, we'll quickly extend that advantage even more. It proved a landmark year both for flying missions and for securing them. Over 30 fresh launches joined the manifest spanning Electron and HASTE. Those originated from a well-balanced set of clients reaching across the U.S., NASC and defense, commercial constellations and international organizations. Numerous repeat buyers signed fresh agreements, frequently involving bulk purchases and several launches, yet we brought in new names as well, showing that our small launch clientele keeps on growing. During Q4 by itself, we closed a new multi-launch arrangement with BlackSky covering 4 additional launches, raising the overall count of missions they've reserved with us to 17. We likewise finalized a deal with a fresh confidential client backing national security. As ever, our pipeline for Electron and HASTE stays robust, and we're eager to keep signing new and novel missions along with standard repeat and mission profiles in 2026; moving on now to Space Systems. Rocket Lab isn't new to serving as a prime contractor, yet in Q4, we issued an announcement underscoring our considerable expansion in the satellite market and further solidifying our standing as a preferred disruptive prime. The Space Development Agency — SDA — awarded us a contract worth $816 million. Under it, we'll build an advanced constellation made up of 18 spacecraft, each equipped with sophisticated sensors for missile warning, tracking, and defense. These are meant to provide detection and tracking of emerging missile threats on a global, persistent basis. That makes it the largest single contract ever booked by Rocket Labs. On top of that, since we're a top merchant supplier to the other Tranche III prime contractors, there are additional subsystem chances that could bring total capture value to around $1 billion, through delivering payloads, solar power reaction wheels, star trackers software, and other offerings drawn from our wide-ranging capability portfolio. One thing worth stressing: bringing Geost into the fold played a major role in securing this award. Rocket Lab is the only commercial company that designs and manufactures both the satellite and its payloads internally for the Space Development Agency and the tracking layer Tranche III, which supports the government's goals of speed, resilience, and affordability in missile defense based in space. This award follows our previous win as prime contractor for SDA's transport layer Beta Tranche II program. Taken together, the two programs mean we now hold more than $1.3 billion in signed SDA contracts. What I'd highlight about what we just announced isn't just that we received a large contract. It's the trend: Rocket Lab continues to win significant awards that once went exclusively to the legacy aerospace primes — those old-guard players. What we're watching is a fresh order taking shape across the defense sector, with firms such as Anduril and Palantir stepping up as key forces shaking up the sluggish, oversized incumbents. In space, Rocket Lab is plainly doing the same thing and pushing the old guard aside. Alright, moving to Mars. During Q4, the ESCAPADE mission got off the ground, and the two satellites we constructed for NASA and UC Berkeley are now traveling steadily toward the red planet. Through ESCAPADE, we've demonstrated that decade-class missions can be delivered on timelines that are dramatically compressed and with budgets far smaller than what interplanetary missions normally require. The reason we were able to accomplish this comes down to vertical integration, which gave us strict control over the timeline and the costs. At this point, both of the spacecraft have been successfully commissioned and are positioned in a Loiter trajectory close to L2 — a Lagrange point located about 1.5 million kilometers from Earth — and soon our primary role in the mission will come to an end, with the UC Berkeley team taking over control next month. Once that transition occurs, we will still be chairing Blue and Gold when the two arrive at Mars orbit in September of next year. It's possible that our efforts on ESCAPADE already qualify as a mission success, but when it comes to Mars, there's still work left undone. We've never hidden our view that Rocket Lab holds the strongest position to carry out NASA's Mars telecommunication Orbit program.
An NTO is going to be the cornerstone for all subsequent activity on Mars, opening up science today and future human exploration; a rare blend of flight-proven spacecraft, extensive deep space mission know-how, dependable launch vehicles, and full end-to-end space systems as a vertically integrated mission provider will let us deliver this. Some of the most ambitious and successful missions to Mars ever flown have been made possible by our hardware and software, such as the Mars Insight Lander, Perseverance Rover, Ingenuity Helicopter — Mars runs through our veins, and these days Rocket Lab has more hardware both on and orbiting Mars than almost any other firm. Alright. Moving to programs. LOXSAT, our launch plus spacecraft effort to construct and place an on-orbit cryogenic fuel depot for NASA, hit a key milestone. That spacecraft is finished, and we'll be moving steadily toward a launch later this year. Alright; there's also an exciting update from our Space Solar division. Some context on where space solar power stands today is needed, so please indulge me for a moment. The satellite sector is growing fast and is expected to expand 7x by 2035. Every one of those satellites will require solar power. Rocket Lab leads the world in space solar power. So it shouldn't shock anyone that we're positioned best to serve this expanding market; beyond that, bold possibilities are emerging from space-based data centers. As AI and compute needs pushed terrestrial data centers to their limits, firms are starting to seriously consider relocating data centers into orbit, where cool conditions and endless solar energy can be harnessed. Yet growth at this rapid pace, for both standard constellations and forward-looking endeavors such as space-based data centers, will be held back if conventional solar cells remain the sole choice. So against that background, I'm delighted to announce that Rocket Lab is bringing out space-optimized silicon solar arrays. Silicon has been used in space before, but it has always been plagued by poor radiation tolerance, extremely short lifespans, and weak performance. Our team are the space solar specialists, having created some of the most sophisticated cells for flagship missions to the Sun and the majority of today's Mars missions. What this team has built is a silicon array that changes the game. Through the use of silicon, we can achieve a genuinely low cost per watt at industrial scale, making possible gigawatt class power generation in space at kilometer size scale via mass manufacturable, lightweight, modular systems; we've additionally gone a step further by creating a hybrid solar array solution combining high-efficiency cells with silicon cells, an approach that draws on the strengths of both technologies. Where size, weight and power or performance matter most, traditional high-efficiency cells are what makes it possible. Where cost schedule or cost constellation scale is what's needed, silicon cells can rise to the occasion. Balancing and trading off those considerations is exactly what hybrid arrays make possible — they bring the two together to hit peak performance at an attractive price point. That brings us to new products, and then to new acquisitions. Speaking of acquisitions, the thing everyone clearly wants to hear about is Mynaric. Regulators in Germany continue to move carefully and systematically through their review. While that process plays out the way we anticipated, there isn't a great deal more we can share right now. Once it wraps up, though, we're eager to bring everyone up to speed. You'll find a handful of narratives circulating in the press offering varying takes on where the deal stands. My advice would be: take what you see in the media and on the internet with a grain of salt. Separately, this month Optical Support, Inc. became part of the Rocket Lab team. OSI, headquartered in Tucson, is a leading designer and manufacturer of bespoke high-precision optical and electro-optical mechanical instruments. For national security and commercial satellites, OSI's technology serves as a crucial enabler; within Rocket Lab's payloads for space protection, space domain awareness, missile warning and tracking defense, they represent an essential subsystem. The chances for vertical integration here are obvious, and we're eager to grow production and capabilities to support our customers and our own programs, just as we've managed with numerous other acquisitions that worked out well. Finally, we've also brought Precision Components Limited of New Zealand into the fold — once again, a supplier we already knew and trusted, now part of the family; through this deal, we've set up a new precision machining complex that delivers an enormous boost to machining capacity. It's worth pausing briefly to consider how strategically important our recent optics-focused purchases are. Combining high-performance RF and optical payload technologies in-house opens up lucrative possibilities with national security and commercial customers. Programs such as Golden Dome and other proliferated mission architectures depend on them. Controlling the payload chain makes possible — it makes possible standout performance along with tighter control over schedule, cost and, above all, high-volume constellations. This approach has already proven itself through the SDA Tranche III award, and we anticipate it will generate still more value and opportunities for us this year and afterward. We got yet another strong endorsement of our capacity to deliver vital national security and defense programs when the NDA recently picked us for Shield. Essentially, we've now been onboarded into the program, whose contract value reaches as high as $151 billion, opening the door for us to bid on future launch and space systems contracts that bring these capabilities to the war fighter with greater agility. Everything I've described comes down to a single point: Rocket Lab is a disruptive leader in building the future for space and defense; that message was reinforced by a recent visit to our Long Beach facilities from the Secretary of War, Pete Hegseth, as part of the arsenal of Freedom Tour. That visit really showed the vital assistance we're providing to the war fighter right now, and it demonstrated our ability to handle needs that keep changing down the road. And finally, before Adam gets into the financial details, here's where things stand with Neutron. There's a great deal of progress to report on Neutron, though I'll begin with what I'm certain is top of mind for everyone: the Stage 1 tank update. Back in January, we told you that Neutron's Stage 1 tank burst while undergoing a hydrostatic pressure test at the Space Systems complex located in Middle River. Look, failures happen fairly often while a rocket development program is going through qualification, but I want to emphasize that this one caught us off guard; in the end, we had expected this tank to make it through qualification. The tank did reach the flight loads we had anticipated, but when we got ready to open up the test bound and drive the pressures and loads past that point so we could understand the structural margins, it gave way sooner than we had anticipated. Our post-test review process found that a manufacturing defect caused a loss of strength at a critical joint within the structure — more specifically, in the area of the tank closeout, which is an autoclave produced part that connects with the bulk composite laminate of the tank and the <<inaudible>>.
Looking at the hardware and the test data, the review indicated the tank did what we expected it to do in other respects. The initial tank was laid up by hand through a third-party contractor during the period when we were bringing the automated fiber placement machine online. And that handlaid approach is where the defect came in. The choice to engage a third-party contractor came down to schedule in the end, since it let us get the first tank built quickly while at the same time bringing the AFP machine into service for tanks to come. Running parallel development paths in this way isn't unusual for us — it helps compress timelines, and it can be an economical method for iterating on prototypes and first articles while simultaneously establishing long-term production capability so we can scale quickly later on. The following tank is in production already. This one is being manufactured on the AFP machine, which removes any chance of that hand defect appearing again. It deserves noting that Neutron's second stage was largely produced — was done completely in-house and passed qualification — apologies — it deserves noting that Neutron's second stage was built fully in-house and cleared qualification with room to spare. In addition to the manufacturing process change, we're also applying a few small design modifications to the first stage tank aimed at adding margin and making it easier to build. To be clear, the overall tank design pleases us. But given that we're building a new one, we figured we should seize the chance to adjust a few things and get it optimized. When it's finished, the new tank is going to go through a thorough test and qualification campaign to confirm flight readiness, and we intend to give that process the time it requires. Bringing a dependable rocket to market will always come first, even when that means a few additional months. In the end, the new tank together with the production design tweaks and the test and qualification campaign will shift Neutron's time frame somewhat. Accordingly, Neutron's first launch is now targeted for Q4 2026. Neutron remains on track to reach the market on an extremely aggressive timeline. What's more, the vehicle we bring to the pad will be robust and thoroughly tested. We're eager to share further development progress as we move through the last development phases this year. Okay. Moving on to some Neutron program milestones from the past quarter. The next few slides will show you why I'm calling this the quarter of qualification. We made enormous progress in Q4 and in Q1 so far, successfully qualifying critical flight hardware all the way from large structures down to component level systems; during Q4, the Hungry Hippo fairing passed qualification successfully and then in Q1, it arrived at wallops; it's an exciting period in Virginia as Neutron flight hardware begins to show up and we can move into the final assembly and integration and test phase. Regarding the Hungary Hippo in particular, what remains is work on fluid systems, fitting canards, adding thermal protection systems, and naturally, full end-to-end testing. As we move through those steps ahead of the inaugural flight, a second Hungary Hipo is already being built for the following Neutron launch vehicle. Another qualification we can now check off is Neutron's thrust structure. That piece of Neutron is genuinely intricate. It has to hold up under 2.1 million pounds of thrust, which is equivalent to more than 44 electrons lifting off at once, just to give everyone some perspective. That structure has officially entered final integration, the last barrier standing between us and integrated system checkouts, cryogenic proof tests, vehicle hot fires, wet dress, and of course, launch. Before it ships out to LC-3, it will pass through avionics, fluids, and subcomponent integration. In the meantime, over at Middle River, Neutron's interstage is going through a qualification campaign of its own prior to shipment to LC-3. During flight, Neutron's second stage sits inside it, then travels through the mouth of the Hungary Hipo on its way to orbit. As with the Hungary Hipo, the interstage stays connected to the first stage and gets reused. That means it has to endure a rigorous test program so we can be confident it can handle the forces of launch and landing repeatedly. Stage 2, meanwhile, is in final integration and being prepped for its debut on the test stand at LC-3. That's a purpose-built rig sitting atop the LC-3 launch mount, where we'll run a batch of integrated tests before finally proceeding to hot fires on the stand; that will be L3's first experience of an Archimedes engine and a massive milestone for the development program; we're eager to test it soon. That leads me to the last but certainly not least item, Archimedes. At this moment, the engines are in boot camp; we have shown them no mercy whatsoever; testing engines within expected limits is all fine and good; but from experience, I've come to learn that space flight tends to throw unexpected things your way. Rocket engines usually don't fail when everything is dull and you can lean on analysis and simulation to bound and then fully grasp performance. In the end, engine reliability comes through testing. There is simply no alternative; so that's exactly what we're doing, and we're really driving them through the edge cases, pulling inlet pressure way down, inducing cavitation and broadly subjecting them to genuinely brutal treatment. In the end, you need to understand how the engines will behave across a really broad set of scenarios on the ground before you send them into the air and discover it during flight. Far too many rocket companies have skipped this, and it usually doesn't end well. This mirrors the very process we followed while developing Rutherford, the engine on Electron. At this moment, we have over 800 of those engines successfully reaching space. So the same reliability and rigor will be applied to Archimedes. In addition to the Stage 1 tank, we enjoyed a very favorable quarter regarding Neutron progress, and what you're seeing here is a snapshot of the extent of advancement we've achieved and delivered on the road to first launch. Qualification is being passed by major structures and subsystems; and for the first time, hardware and final integration are in hand. These represent the last steps prior to entering integrated testing at the pad with hot fire stage tests, followed by wet dress and then, naturally, launch. Past the vehicle itself, all the supporting infrastructure needed to facilitate first launch and beyond has been put in place. OC3 has clearly stood up, plus production and test facilities are all humming, while the regulatory work is tracking along exactly as we expect. What to watch for over the coming months as evidence that we're marching steadily toward launch includes additional hardware making its way to the launch site, we'll be performing extensive testing of flight hardware, and then obviously, that will lead up to Neutron's first flight. That concludes the operational highlights; so I'll turn things over to Adam for the financial overview and outlook.
Thanks, Pete. Revenue for the fourth quarter of 2025 hit a record $180 million, landing at the top end of the guidance range we'd given previously and marking a year-over-year increase of 36%, which is impressive. Both of our business segments contributed significantly, and that drove this strong performance. Looking at it sequentially, revenue rose by 16%, which highlights the momentum that continues across the business. In the quarter, our Space Systems segment brought in $103.8 million in revenue, a sequential drop of 9.1%. The drop was driven primarily by our Satellite Platforms segment as well as our Solar operations, and even with the programmatic unevenness in revenue recognition that ASC 606 introduces and the related subcontractor progress that shows up now and then, both are still doing extremely well. Fortunately, our expanding diversification across Space Systems and Launch frequently allows us to produce steadier top line growth even when individual product lines swing around; this quarter was one of those cases, where Launch Services strength outweighed the Space Systems declines, bringing in $75.9 million in revenue — an 85% quarter-over-quarter rise, reflecting an increase from 4 to 7 launches in the period, which included 1 HASTE mission. Looking at the whole year, we brought in $602 million in revenue during 2025, which works out to a strong 38% increase compared with the prior year. Moving on to gross margin now. For the fourth quarter, GAAP gross margin came in at 38%, right in the middle of the 37% to 39% range we had guided toward, and that marks a 100 basis point improvement from the third quarter. On a non-GAAP basis, gross margin for Q4 reached 44.3%, which likewise matched our earlier guidance of 43% to 45% and represented a 240 basis point gain quarter-over-quarter. Better gross margins quarter over quarter came mostly from two things: Electron absorbing more of its fixed costs thanks to a busier launch schedule during the period, plus a bigger lift from our Space Systems components operations, which carry higher margins. Looking at the whole year, gross margin on a GAAP basis reached 34.4%, up 780 basis points from the year before, and non-GAAP gross margin came in at 39.7%, which is 770 basis points higher year-over-year. On a related note, production-related headcount stood at 1,244 when Q4 closed, a gain of 46 versus the previous quarter. Before we turn to backlog, I'd like to pause and step back to give some context on how far we've come against our long-term financial model since listing on NASDAQ in 2021. Top-line sales are up close to tenfold, which works out to a compound annual growth rate above 76%. Each year our gross margins have risen, and the amount we keep from every dollar of revenue has more than doubled. That kind of improvement speaks to how disruptive and strong our competitive standing is within the industry, and to how much customers value the differentiated products and services we offer throughout the company. When you pair that sales growth with these better margins, the business now sits on a sturdy base and is on track to deliver substantial operating leverage along with cash flow generation over the long run. One last thing: I wanted to highlight SG&A as a share of revenue, since it's encouraging to watch that figure keep declining as we bring the business to greater scale. We keep pushing this business to operate with real intensity, and my view is that we're set up to generate additional expansion and greater efficiency during 2026 and afterward. Moving on to backlog now. We closed out Q4 2025 with roughly $1.85 billion of total backlog, which is a remarkable 69% sequential increase, mostly attributable to our recent SDA Tranche III tracking contract win, which was announced last December. As we've said previously, Space Systems backlog specifically can be uneven because of the timing around these progressively bigger needle-moving program opportunities. Yet after they're awarded, they can substantially reduce risk on revenue growth for multiple years. Our pipeline continues to grow strong, covering multiple launch agreements for Electron, HASTE, and Neutron, along with significant satellite platform deals in both government and commercial sectors. Right now, Launch backlog represents around 26%, while Space Systems makes up roughly 74%. Going ahead, we expect approximately 37% of the backlog we currently have to convert into revenue over the next 12 months, and that figure includes our initial Tranche III revenue recognition assumptions, which we believe will prove conservative and, combined with a solid sales pipeline, ought to drive further top line gains beyond the conversion we've already factored in for the current 12-month period. Now turning to operating expenses. For the fourth quarter of 2025, our GAAP operating expenses were $119.3 million, which landed below the $122 million to $128 million guidance range we had given. Non-GAAP operating expenses for Q4 came to $104.5 million, also finishing under our $107 million to $113 million guidance range. When you look at why both GAAP and non-GAAP operating expenses went up from the prior quarter, the main driver was continued expansion in prototype spending and additional headcount to back our Neutron development program. To put a finer point on it, we stepped up investments in propulsion as Archimedes engine testing continues, alongside testing and integration work on mechanical and composite structures at our Middle River, Maryland facility. If you focus on R&D alone, GAAP expenses rose $8.1 million versus the prior quarter, and non-GAAP expenses increased $7.7 million. As I mentioned earlier, the growth came from scaling up composites manufacturing and testing, along with higher spending on composite structures and fluids. R&D head count at the close of Q4 was 1,012, down 7 from the previous quarter. Turning to SG&A, GAAP expenses declined by $5.1 million compared with the prior quarter, and non-GAAP expenses were down $1.3 million quarter-over-quarter. The main drivers of those decreases were reduced legal and other professional services fees related to transactions involving M&A and capital markets activity, plus a slight cutback in marketing spending. SG&A head count ended Q4 at 389, which is 4 more than the prior quarter. In summary, we closed the fourth quarter with 2,645 employees, which is 43 more than we had at the end of the previous quarter. Turning to our cash position: during Q4 of 2025, we spent $49.7 million on property, equipment, and capitalized software licenses. Compared with the $45.9 million we reported in Q3, that represents a $3.8 million increase. This higher figure reflects our ongoing investment in Neutron development, including continued testing and integration work at LC-3 in Wallops, Virginia, and in Middle River, Maryland, as well as expanding the capabilities of our engine development facility in Long Beach, California, and building the return on investment recovery barge in Louisiana. As Neutron's first launch draws nearer, we expect capital expenditures to remain elevated as we put money toward testing, ramp up manufacturing, and build out infrastructure. In Q4, GAAP EPS was a loss of $0.09 per share, compared with a loss of $0.03 per share during Q3. The bigger GAAP EPS loss from one quarter to the next is mostly tied to the $41 million tax benefit recorded in Q3, which came from partially releasing the valuation allowance on our corporate deferred tax assets, and that release resulted from picking up an equivalent amount of deferred tax liabilities tied to purchase price accounting for the Geost acquisition. GAAP operating cash flow showed a use of $64.5 million during the fourth quarter of 2025, against $23.5 million in the third quarter. That $41 million sequential jump in cash used was nearly all attributable to when employee equity program related tax payments landed.
Continuing the same pattern we saw with capex, our cash burn is going to stay on the high side, and that's because of work on Neutron, the extended lead times we're dealing with for SDA, money going into producing later Neutron tails, and building out infrastructure so we can grow past that first test flight. Taking it all together, non-GAAP free cash flow — which we calculate as GAAP operating cash flow minus what we spend on property, equipment, and capitalized software — came in for Q4 2025 as an outflow of $114.2 million, versus an outflow of $69.4 million in Q3. We closed the fourth quarter with $1.1 billion across cash, cash equivalents, restricted cash, and marketable securities. That quarter-over-quarter bump in liquidity came mostly from money raised selling our common stock through the at-the-market equity offering program, which brought in $280.6 million during the quarter. The main purpose of those funds is to back acquisitions — like the pending Mynaric deal we've announced, the Optical Support, Inc. and Precision Components Limited deals we just closed, and other candidates sitting in our strong M&A pipeline — plus ordinary corporate costs and working capital; we finished Q4 well positioned to pursue both organic and inorganic growth, deepen our vertical integration of the supply chain, broaden strategic capabilities, and enlarge our addressable market, just as we've successfully managed to do before. Our adjusted EBITDA loss for Q4 2025 totaled $17.4 million, which came in under the $23 million to $29 million loss we had guided to. That $8.9 million sequential narrowing of the adjusted EBITDA loss came from substantial gains in revenue and gross margin, partly offset by higher operating expenses tied to Neutron development. That brings us to guidance for Q1 2026. For the first quarter, we're looking at revenue somewhere between $185 million and $200 million, which works out to 7% growth quarter-on-quarter at the midpoint and 57% growth versus the same quarter last year. We're anticipating a slight decline in gross margins on both a GAAP and non-GAAP basis, with GAAP gross margin landing between 34% and 36% and non-GAAP gross margin between 9% and 41%, reflecting a modest drop sequentially as Space Systems makes up a bigger share of the mix relative to higher-margin Launch, and as margins within Space Systems get less favorable. For Q1, GAAP operating expenses should fall between $120 million and $126 million, while non-GAAP operating expenses should fall between $106 million and $112 million. Those sequential increases mostly reflect continued Neutron development and Flight 1-related spending, including staff costs, prototyping, and materials. That said, we expect spending to shift out of R&D and into flight and inventory as 2026 progresses, which is an encouraging signal of progress as we get closer to Neutron's transfer to flight and, as a result, to adjusted EBITDA positivity. I'm optimistic, given the impressive strides we've made toward this milestone, and we currently expect Q1 to mark peak Neutron R&D spending. For the first quarter, we anticipate GAAP and non-GAAP net interest income of $8 million, driven by increased cash balances and the conversion of roughly $117 million in convertible notes since December 31. We expect our first quarter adjusted EBITDA loss to fall somewhere between $21 million and $27 million, with basic weighted average common shares outstanding at about 605 million shares. That figure includes approximately 46 million convertible preferred shares and accounts for the conversion of roughly 23 million shares from our outstanding convertible notes so far in Q1. At this point, only 7.5 million shares remain outstanding, or 11% of the original $355 million issuance. And considering additionally the retirement of the Trinity equipment line in Q4, we have substantially eliminated — we have eliminated indebtedness from the business. Finally, as in prior quarters, we anticipate negative non-GAAP free cash flow in the first quarter will stay at elevated levels, due to ongoing investments in Neutron development and scaling production. This does not include any potential offsetting effects from financing activities. Last but not least, here are some of the upcoming investor events that we'll be attending in the next few months. And with that, we'll hand the call over to the operator for questions.
Adam, perhaps I'd like to begin with the backlog. I'm curious whether you could help us dig into it a little more; and perhaps remind us what's in there — does it cover the 40% of revenue from SDA Tranche II, 10% of maybe Tranche III? And what are you counting from Neutron and Electron here?
I'm not sure how much of that you heard — so the — every SDA contract got added into backlog; so whatever is left for SDA Tranche II transport layer is still sitting in backlog. Clearly, what has been recognized as revenue isn't in there anymore. By the close of Q4, none of the Tranche III contract awards had been recognized; so that entire value is in backlog right now, and it will begin converting into revenue and exiting backlog, obviously, through that process. When it comes to Neutron, I believe we've said previously that we have a number of flights that are representative within our Launch backlog as reflected in our filings. So with luck, that addresses your question about how backlog is composed.
Yep, that helps. Maybe one more thing on top of that. Turning to Neutron — now that the first flight has moved to Q4, what's the right way to think about cadence? Do you still expect around 3 launches in the first 12 months following that initial one? How much confidence do you have in getting the second tank built, and should we anticipate any increase in CapEx now that the second tank is in production?
I can take a few of those, and perhaps you and summer can jump in too. About the tank, Andres — I believe everyone's clear on what has to happen there. A significant portion of the second stage tank was already produced on the machine, so that takes care of the issue. As for how to think about the subsequent flights, the situation isn't as grim as pushing every one of them out by a year or tying them to the first flight, since, as the presentation showed, we're already working at full capacity on more Neutron tail numbers. Because that tank and the AFP machine let us produce a tank far quicker than the hand lay method can, the shift onto later flights should come together a bit more quickly — nothing else in the qualified hardware has been stopped. So my sense is we'll be in a stronger position on that front.
Yeah. So Andres, on your question about capital expenditures and similar items tied to that second tank, the one taking the place of tank number 1 that burst — what I'd say is the upside right now, like Pete mentioned, of having the AFP in place is that we can not just build it more quickly, but the real expense of making that second tank comes in pretty low. Tank number 1 ran us a lot because, as Pete noted before, it was built by hand lay-up; that consumed a great deal of time, whereas this one will go much faster. On top of that, given that the AFP is now commissioned, essentially what we're discussing is variable costs for tank materials, primarily, since the labor we already have is basically built into the model. So no additional CapEx will come from this, and when it comes to R&D, the effect of the tank failing is really not — the tank itself isn't all that material.
I'd like to ask about data centers in space. I believe you mentioned there's a lot of interest there. Obviously this has turned into a hot topic across the industry. Could you help us understand how those early conversations are progressing with possible customers who want to pursue this? And is it reasonable to expect some kind of Rocket Lab content inside a space data center, say, over the next 2 or 3 years?
Thanks for asking. Look, when it comes to data centers, we're at an early stage. If you examine some of the models out there, several factors need to come into focus before they turn into the sensible option compared with terrestrial; however, we never want to let an opportunity slip by, and for a while now we've been working on the silicon arrays and power solutions, with our attention on mega constellations and high-volume power applications. But if you step back objectively and consider all the difficulties involved with placing data centers in orbit, it really comes down to 3 things. First is the cost and cadence of Launch so the model can close. Second is heat rejection through various means; and third is simply sheer power, meaning a gigawatt of electricity, electrical power. So solar arrays of multi kilometers in scale are what's needed; so we wanted to ensure that whether they leave this Earth or not, there'll be Rocket Lab logos all over that stuff. So to the best of my knowledge, nobody else has a silicon solution quite like what we've developed.
Got it. And regarding what you said about heat rejection — I suppose the rate eater — is that something you can do internally that will take time to build up? Or does it come from outside? I'm just wondering what technically has to be worked out there.
Yeah. Well, obviously every one of those spacecraft has radiators — you make heat, so you've got to get rid of it. There are different approaches to moving heat through a spacecraft's radiator. So to me that isn't some enormous engineering hurdle; it's simply that the scale needed is something nobody has managed before. That's where the difficulty lies. Still, let me be clear — I don't see us constructing enormous AI data centers anytime soon, but for those at least trying it out and wanting to head in that direction, I believe we offer plenty of strong solutions.
Understood. If I may squeeze in just one more brief question. Regarding those conversations, could you provide some insight into what kinds of customers they involve? Are they, say, newcomers, or perhaps unconventional buyers who are sort of looking into this concept alongside you?
Yeah. I should mention we need to tread somewhat carefully on this, but I'd put it this way: there are definitely a greater number of nonstandard parties examining a solution of this type compared with conventional ones.
Alex Preston here, filling in for Ron. You touched on how things are going with the Mynaric acquisition, but what I'm really curious about is the bigger picture in Europe — the broader environment, if you will. There appears to be increasing demand for what you'd call homegrown launch and national security space capabilities. Do you and the team notice that trend, and where do you think it's heading? Pete brought up that no other small launch provider has truly pulled it off over the past year, yet plenty of people still see it as a priority, I believe.
Alex, appreciate that question. One big reason we're drawn to Mynaric, and why we view it as significant, is precisely that point about Europe and Europe more broadly — plenty of spacefaring nations in that region possess minimal capability paired with enormous ambitions and very tight schedules. Everyone always wants to develop domestic capabilities, of course. But here's the truth: if your goal is to get these sorts of capabilities up and running incredibly fast, you simply don't get the decades required to often construct these sovereign capabilities. The equipment, the facilities, and equally the intellectual property and knowledge tend to be highly specialized. That's why Europe looks like a tremendous opportunity to us, a genuine expansion beachhead — we're able to deliver solutions at the component level. We can deliver solutions for a complete satellite system. We can deliver launch, and you've even seen European space agencies buying Launch from us at this point. And the moment we have a footprint inside Europe proper, becoming eligible to take part in your European programs turns possible. So my view is — it's a tremendous opportunity. Right now there are literally billions and billions of dollars of well-funded government programs in motion, and the time lines tied to them are conducive — or, I'd say, not necessarily always conducive — to establishing sovereign capability.
Understood. Then, based on your comments, would it be fair to say the stance remains largely positive, in contrast to Europe perhaps beginning to pull back somewhat from providers headquartered in the U.S.?
No, I'd describe it as highly positive. Obviously, Europe wants to build its own independent capacity, yet in many of the discussions we've had, people there are quite practical and clear-eyed that developing that capacity requires a great deal of time. So partnering with something like a Rocket Lab Europe is an excellent path forward.
And just very briefly, would you say that’s the same situation at launch as what happened with Space Systems, where I believe Europe already has somewhat more indigenous capability?
Yes, they are definitely making a serious effort, though I’d say they aren’t achieving huge success — yet that’s simply how hard launch is; still, I believe launch matters so much strategically. You can build as many satellites as you like, but if you’re unable to get them all up, then it’s fairly meaningless. That’s why the European Union and ESA have launch vehicles that, on the surface, aren’t especially commercially competitive, yet they will never disappear, since those nations require access to orbit. So I’d anticipate that continuing for a while, along with ongoing investment being made in — into Launch for — for Europe. That said, everybody is pragmatic, and if you have to get things all up, then you call.
Yeah. Perhaps we could return to Neutron for a moment. I'm grateful for the cadence update, and it seems that given the delay, you'll obviously keep expanding certain capabilities within the Neutron infrastructure itself. But following the test flight — we're thinking roughly Q4, and should it land late in Q4, I'm unsure of the exact timing — what's your view on when that initial revenue-generating flight might occur? What timing do you foresee there? And when you consider that, it likely requires a greater degree of reliability. Given all that, do you still intend this to be a recovery mission?
Erik, I appreciate the question. The schedule for Flight 2 is always going to hinge on what comes out of Flight 1. Should Flight 1 proceed without a hitch, then when it comes to getting the second vehicle onto the pad, we'll push to make that interval as brief as we can. If there are items needing correction, then there are items needing correction. Still, under nominal circumstances, the schedule stays in line with what we've previously described. And the vehicle is going to be fitted with every one of its requirements for Flight 1, even for a downrange landing; we'll try the reentry and the landing, bearing the stress as it comes down. Again, assuming everything proceeds favorably, then on the following one we'd plan to position a barge beneath it. If we instead guide it into the sea, then we'd likely move to Flight 2 and achieve that gentle touchdown before we proceed to place infrastructure beneath it, which could prove expensive for them should we cause damage.
Thanks. Turning to Electron for a moment — you ran a strong campaign in 2025 with 21 launches that went off successfully. So what are you seeing for this year, based on the manifest and your internal plans? And could you also speak to how the mix breaks down between standard Electron flights and HASTE?
Sure. I'm not certain exactly how much detail we've already put out there on that. That said, we definitely expect to fly more this year than we did last year. The bookings and manifest are really filling up, as you've seen, and right now we're turning out Electrons roughly every 11 to 13 days. So things are going very well on that front. That said, I'll hand it to Adam in case he wants to weigh in — do you want to walk through the schedule for the year?
Right, Erik. So, in line with what we've said before, we're seeing solid expansion potential within Electron. And by Electron, I'm referring to both Electron and HASTE. So you should anticipate a rise in regular Electron missions as well as expansion on the HASTE side of things. Typically we steer folks toward roughly 20% growth, which I'd call a fairly sensible projection for how this business grows across the near and intermediate, and perhaps even long, horizon. So I'd say we've definitely instructed the production team to build considerably more rockets during 2026 compared with 2025. And like Pete said earlier on this call, we booked more than 30 Electron launches in '25; and we always pick up turns orders. So look, if you basically assume a 20% increase in the Launch business, leaving out Neutron obviously, I think that's likely a pretty solid position to land on.
Peter, perhaps I'll start with you. Your prepared comments about the OSI acquisition gave the impression that it was helping you even more with the customer, in terms of how appealing your services and capabilities are, even though the announcement suggests OSI was already among Geost's suppliers. So is the customer really focused then—can we take it—just on the vertical integration side? Or are there also capabilities, functionality, features from that acquisition from a systems standpoint that are appealing too? I'm just trying to get a sense of how you believe customers are actually viewing this, if that makes sense.
Yes. No, that's a wonderful question, Trevor. To be honest, most customers likely aren't too concerned about that detail. What really matters to them is getting this sensor delivered on time, with a price and performance level unlike anything they've seen. That's exactly what we deliver. To make that promise possible, we need the most critical piece of many of these optical systems: the optics themselves. By bringing optics in-house and owning that capability, we gain real certainty around cost, schedule, and innovation. And yes, that company definitely was a supplier to Geost. Once we acquired the Geost business, we immediately met with its leadership group and asked: which parts of the supply chain could become bottlenecks and stop us from providing truly disruptive, affordable components or programs to our customers? This topped the list. In my view, that sets us apart from other payload providers that buy optics from outside vendors. It's also the priciest component and the one with the longest lead time in any of these explicit optical payloads. So owning it mattered.
Great, that's really useful. Adam, perhaps a fast follow-up question for you. Regarding what you said in your prepared remarks about the backlog and how Tranche III will go — it sounds like it might be cautious regarding what will be recognized during that initial 12-month span. Could you perhaps take us through some of the pluses and minuses of how — what is, I suppose, driving that Tranche III rev-rec? Is it merely customer scheduling for when they'd like deliverables? What's the — just give us perhaps a bit more detail, that would be great.
Yeah; so I believe we've explained before that usually when you secure one of these programs, you're able to recognize revenue sort of around 10% during the first 12 months following award, then 40% during the next 12 months, 40% during the third 12 months, and in the final 12 months, it's roughly another 10%. So you end up with a fairly typical bell curve. What I'd say is that with — what truly limits our capacity to move more quickly is actually our subcon deliveries, correct? So what would really either help us speed up and get through those gates and milestones and rev-rec faster is our subcons' capacity to deliver promptly. And so I think that all ties back to what Pete mentioned earlier and the significance of vertical integration. So if we can simply own more of the platform, we gain greater control. And that lets us have more predictability in how we time revenue recognition and so on; so I'd say a major task for us in 2026 across our engineering and production teams is to truly ensure we remain on top of which parts are still sourced from third parties, ensure that they remain on their deliverables so we can, again, get the program accelerated as much as possible and get more of that revenue recognized. So again, we enter it fairly cautiously. I think what we've — what — if you examine the pure conversion at 37%, I think that was brought up earlier regarding backlog converting, I mean, clearly, a share of that is Launch but the share tied to Space Systems. Some of that originates from the components and subsystems entirely unrelated to SDA Tranche II and Tranche III; but what's included there for Tranche III is, again, presuming some fairly cautious delivery dates from our subcons, and ideally, we can collaborate with them to do better.
Somehow we've got Andres here now — no idea what went on there — but it's all fine. I had a question about Space Systems. At first glance, it appears to have landed somewhat below what consensus was probably looking for. So I'd simply like to understand the pluses and minuses involved. You did walk through it, I realize, but what could have caused consensus to get a bit ahead of itself this quarter?
Sure. I'm not convinced consensus is all that skilled at separating out the different components of the business, or even at drawing much of a distinction between Launch and Space Systems. Beyond that, inside Space Systems, I doubt they truly examine how our platforms operation compares with the subsystems operation. So, as I noted during my prepared remarks, one point is that it's tough to — well, you simply can't — insofar as you're able to steer how execution unfolds for your rev-rec obligations under ASC 606, everything hinges on how effectively your subs perform, correct? Plus how closely you collaborate with them to keep them on schedule. Doing everything within your power, I think we've all witnessed in some rather public settings customers involved in these programs discussing how the supply chain has hit some snags, including from certain sources, say, for instance, the optical terminal providers. So, regarding what we do, we're constantly seeking methods, as Pete said, to lessen any reliance on outside parties to the greatest extent possible. That explains why, if you examine Electron, the degree of vertical integration on that vehicle is high, and Neutron will follow a very comparable path. We're increasingly moving in that direction with what our Space Systems platform offerings provide, where, I'd say, only a small amount remains handed off to outside parties. So once more, it truly comes down to working alongside them and pushing them to deliver as hard as you can, without giving up quality or cost wherever possible. So yeah, I wouldn't put much stock in the level of detail people might have anticipated from our Space Systems business, since 1 of the advantages we now enjoy from having — from operating such a diversified set of businesses, we essentially focus on the top line, on how we can achieve that sequential growth of the business, and at times a larger share will originate from Launch. And at times it will originate from Space Systems, and inside Space Systems, platforms may post a strong quarter while components are weak, and the reverse also happens. And from there it only improves further, and we'll have that many more tools at our disposal once Neutron comes online, which explains why, clearly, getting that first flight off is so important, why we're all looking so forward to that.
Yes. No, that is extremely useful. I suppose, to remain on your topic. I mean, regarding those 2 acquisitions that were recently made public, can you provide any financial details at all about how they performed? And I suppose, how much in costs could potentially be removed because they are now being handled internally?
Yes. What we have in our pipeline tends to be fairly intriguing. It contains a combination of deals that are more impactful financially when it comes to adding revenue and similar things. These specific deals are far more strategic in nature, once again, focused on vertical integration, lowering risk rather than, as I would put it, offering substantial entry to big outside third-party kinds of TAMs, so to speak, or nearby markets. So these are truly more about, I would say, cutting down on some margin stacking and additionally gaining tighter control over the programs. So I would not claim there is a, I would say, significant revenue contribution that will be impactful from the deals we recently announced. Obviously, Mynaric is — would be an entirely different situation should that deal be cleared, since it would bring a considerable backlog along with revenue potential. And once again, our pipeline includes numerous other deals that combine, again, the removal of margin stacking and in certain instances, more substantial revenue contribution as well. But for these 2, I do not believe you need to adjust your models whatsoever for the effect of these 2 fairly minor deals.
I had a question about the Neutron tank failure. Are you guys — do you have high confidence that the manual layup process was the cause, and so the new process won't run into that same anomaly? Or are you still studying what took place?
Yeah. No, we did a full pastry analysis, and we managed to locate the section of tank that triggered the failure. Through analysis, and then through coupon testing too, we were able to reproduce those results. So no, our confidence is very, very high. That value is something we understand extremely well.
All right, that's wonderful to learn. Then you brought up certain sectors where you want greater vertical integration done internally. Could you elaborate on a few of those product categories that could be appealing?
Sure. If you examine the spacecraft landscape nowadays, I'd say the parts where we lack complete control are becoming increasingly fewer. Our RF group is excellent, yet I believe that remains one such field, and we'll aim to strengthen it. We'll also pursue chances to increase scale whenever we can. Still, I suspect this will simply be routine work for us — continually ensuring we aren't hurt by vendors incapable of supplying what we need, while pushing ahead with vertical integration. Yet, as Adam noted, our M&A pipeline is quite packed, with a variety of prospects spanning from matters of this sort that matter, which don't bring large revenue contributions, yet effectively secure revenue since we won't fail to hit milestones. These prospects also extend all the way to genuine needle movers that are far more transformative, and, as Adam likewise mentioned, we always ensure we hold ample capital reserves available to pursue those more significant acquisitions.
Adam, I'd like to dig into backlog — how do you see the chances in front of us across, let's say, the coming 12 to 18 months? Clearly, SDA has been extremely busy. And how do you view what makes up that backlog when it comes to DoD against commercial, specifically looking at the next 12 to 18 months?
Yeah, we're in a lucky position, since government business has typically never been seen as a hockey stick. For us, I think, because we're entering in such a disruptive manner — and we are disruptive, yet also the entire architecture, moving from Geo to Leo, plus the sheer quantity of satellites needed to back that architecture, has simply been so powerful. There are so many factors pushing us along regarding where the opportunities sit. Still, I'd say that on the whole, we have truly large commercial opportunities that we keep pursuing; even so, if someone handed me the option of pursuing a government hockey stick or a commercial hockey stick, I'd choose the government one, because while they might not be as dynamic at times at a program level as commercial ones, they always pay their bills. It's fairly clear cut how you deal with them; and within that government market, the folks we're up against appear to be competing with their hands tied behind their backs, correct? That means we act far faster; we have many more tools available thanks to our vertical integration. So I like the mix as it leans toward government. I also find it quite reassuring to have that large commercial hockey stick opportunity sitting out there too; however, I'd say that the pipeline — when you examine the pipeline of business opportunities, setting aside the M&A side — is a fairly even set of opportunities split between commercial and government. I mean, I'll let Pete give his perspective, but it appears we aren't limited to picking one fork or the other in the road, since we can consider how to capture both. And I believe we've done a solid job balancing, though perhaps Pete would like to comment on that.
Adam, I believe you put that just right. And Mike, yes—there's nothing I could say that would improve on it.
Great. Perhaps a brief follow-up. When you consider Golden Dome, its schedule, and how PWSA slots into that architecture, do you have any fresh perspective on the part you'd play, or on the chances you see once that turns into a genuinely workable commercial opening for you?
Yeah. Golden Dome, I'd call it a fairly intricate thing — clearly it's an enormous program, though much of it sits in classified territory. That makes it hard to say a great deal. Still, on several fronts, my sense is we're nicely placed to capture a solid share, whether that's launch or satellites, optical terminals, many of the optical payloads, the SDA — since the Tranche III SDA award is plainly a missile track payload, an extremely involved pallet, needless to say, and essential to Golden Dome. So as that effort takes shape and keeps expanding, I'd say we're a rather crucial component of that base.
I'd like to pose something with a longer horizon regarding a possible satellite constellation from Rocket Lab down the road, particularly in light of a few announcements lately throughout the industry; and like you noted during the presentation, the substantial expansion in satellites anticipated across the coming decade, has anything shifted in how you're approaching a constellation of your own someday, or which potential applications you might pursue? Or does this remain a longer-term growth opportunity that truly won't rank as a priority until Neutron is flying on a regular basis?
Yes. Appreciate that, Michael. So the main point I want you to walk away with is something I've mentioned before: the lines around this sector are going to become hard to see. Telling apart a space business from a business of another type is going to be a real challenge. And that idea only gets stronger as you watch data centers and the rest of these possibilities grow out there beyond Earth. The way it looks, the really successful enterprises will have fuzzy edges. Would you call them a space company, a telecommunications company, or a data services company? What you're saying holds true right up until Neutron is up and running and we can launch multiple tons while reusing the vehicle. I believe that's the point where we can genuinely focus on rolling out infrastructure. That said, we aren't just relaxing and doing nothing while pondering potential moves. You can probably tell that across virtually every path, we already possess some awareness, pieces, or involvement. By revenue, I mean each sort of chance that might currently be considered or applied in space. So it remains premature, Michael, yet hardly a single day passes without some conversation about it happening internally.
Great. And then perhaps regarding the Stage 1 tank rupture, I'm not sure if I overlooked it, but how quickly are you able to build the second tank now using the new AFP machine? And then will that become even faster as you keep repeating this process over time?
God, look, it's absurd. The AFP machine is simply completely absurd. I can't recall the precise schedule for laying out a dome. But we gauge a dome manufacturer on the AFP in days. Actually, the bigger obstacle for a tank manufacturer isn't really laying up and curing the components; it's joining the various domes and tanks and barrels together, and all the tabs and details of baffles and all those sorts of things are what actually consume the time, but a new tank, we're talking months here not for a complete tank. But from the actual manufacturing of the oral components, it's absurdly fast. And also that, to Adam's point, it's like now that everything is automated, really the sole cost is the raw material that's going in there.
I want to revisit PWSA and hear your big-picture perspective on that program. Given the Geost acquisition, it really looks like your emphasis is going to lean further toward the tracking layer, which is truly striking, so how are you viewing what lies ahead for that within Rocket Lab? Also, we keep hearing about plenty of government reports coming out regarding the transport layer, for instance how challenging would it be for something commercial like Starshield with MILNET to serve as the transport portion. A lot of obstacles appear to exist there, since a commercial Starshield sits at significantly lower altitudes compared with the transport of the tracking layer. That means substantial redesign effort would be needed; I'll leave it there and just ask for your broad take.
Honestly, Jen, we could spend days pulling this apart. So yes, moving up the value chain, so to speak, was deliberate on our part. The transport layers aren't elementary by any stretch — they're an entry point — yet doing the tracking work is an order of magnitude harder and an order of magnitude more valuable. As things evolve for Golden Dome and other sorts of programs, tracking becomes essential; that's where the high value sits, and only a handful of players nationwide can pull it off successfully. Regarding the transport layer disappearing, we've neither heard nor seen anything supporting that. Clearly, there's plenty of chatter about alternative providers. Still, the entire premise of the SDA program is that all the spacecraft are tightly integrated, even when they come from different providers, and there's a common set of requirements everyone has to satisfy for interoperability. So your observation is a fair one. When you've got something quite different, achieving interoperability gets harder. It'll be fascinating to watch how everything plays out. Yet for the objectives SDA is pursuing, to me at least, it makes more sense to have a dedicated transport layer, with the remaining layers — tracking, then custody, and so forth — built on top of it.
That's quite useful. Now, one brief follow-up, if that's okay. I'd like to stay mindful regarding the Mynaric deal and simply let it unfold however it does. But when it comes to optical terminals — and again, I'm hoping this turns out favorably — at what stage might you consider perhaps a different supplier of OCTs? Or could Geost, or the newly acquired OSI, possess some capability you'd eventually leverage to build those optical terminals yourself?
Right. Geost has indeed created certain optical terminals. And clearly, the optics are now ours internally too. Still, this is extraordinarily hard to pull off. When we survey the whole field of optical terminal vendors — and honestly there are merely 3 — Mynaric clearly emerged as by far the finest technologically; these days they're preoccupied with other matters, such as operating their company, yet their terminals are the greatest. So going out and building your own terminal — sure, entirely doable. It's purely a matter of time. And doing so would simply require more time than buying one would.
This is Daniel Hischman, filling in for Jeff Van Rhee. Regarding the Mars Telecommunication Orbiter, that roughly $700 million to $750 million figure, I'm curious — it seems NASA issued an RFP earlier this week for the Mars Telecommunications Network. So the name has shifted somewhat. It sounds like that could involve a multi-satellite architecture, whereas before, only that single orbiter was under consideration. But what can you share about how the competitive landscape and market positioning around this have been developing?
Thanks, Jeff. Excellent question. Yes, so the MTO, as you mentioned, has changed slightly to network. And as additional infrastructure gets constructed on Mars, naturally, a network will need to be established. The MTO was always meant to be the first of water to come. Look, clearly, we believe our position is strong here. There's — we possess the experience. We hold many of the capabilities and many of the demonstrated capabilities, though I think we'll put our best foot forward there. And naturally, others believe they can handle the job as well. That's the wonderful aspect of competition, and we'll find out who wins.
Then Adam, one thing for you regarding gross margins, which clearly are rising dramatically — I believe around 8 points higher in 2025. The Q1 '26 guidance shows them dropping back by a few hundred basis points, and you mentioned the Space System mix shift. Is any part of that mix shift lasting, perhaps because of new business coming online with the SDA transport layer, meaning there will be some ongoing margin pressure? Or in our models should we expect that we return to the more typical pattern of a few hundred bps of expansion once we're in the later half of the year?
Well, gross margin — many factors are at play beneath the surface. As we keep expanding, there was a question earlier from Erik regarding Electron launch cadence, so I noted 20% launch growth in that. If we can exceed that, and I believe there are chances to grow quicker in 2026, that would create a positive upward bias to margin. These bigger, longer-term programs such as SDA Tranche II and Tranche III usually arrive at the relatively low end of our gross margin mix, yet they possess really solid operating margin characteristics, or contribution margin, since there isn't a huge amount of incremental R&D that sits outside of the programs. So I'd say in a quarter where you have much more contribution from the large programs like Tranche II and Tranche III, that will create downward pressure, offset hopefully by growth from an increase in the Electron contributions. The Components business has a fairly interesting range of margins. Some products there are closer to, say, 30 points in non-GAAP gross margin, while others are kind of north of 70 points of non-GAAP gross margin; so there's a wide spread and mix is difficult to predict that far out in the year. I mean I do think there will be a supportive trend towards gross margin, but I think it's hard to really get much granularity beyond, I'd say, maybe 1 or 2 quarters out. But overall, I think as we continue to grow that components mix of the business, more Electron in the mix, it's all going to be positive. I'd say there's one thing to keep in mind: once Neutron starts production, its margins will climb along a path that likely looks a lot like what Electron has shown us, and that has been a really impressive story of margin growth. Still, when you launch a brand-new product such as a rocket, you end up doing things like block upgrades, and those changes help lower your costs and boost performance, which lets you sell more of the rocket's capacity — and that supports ASP and similar metrics. That said, in the Launch business, I'd argue the single biggest factor is rate. Everything comes down to how well you absorb the fixed overhead, or fixed costs, tied to that particular program or product. So here's what I expect you'll see us begin to do: our intention is to provide as much transparency as possible, or to break out the figures between, say, Electron and Neutron once Neutron enters production. That way you can observe that ongoing expansion, with the Electron business running at its model, and then watch the patterns as Neutron scales up toward its own target model. With any luck, reaching the target model — target margins — happens somewhat faster for Neutron, since it's a sensible launch vehicle, though it will still require a number of years. Early on, then, you'll see gross margin that's rather low, perhaps even negative, on some of the first flights. Even so, much like Electron, it should bounce back and turn quite positive quite fast, ultimately hitting the target model. So that was a lengthy response. Once again, I believe the trends do back gross margin expansion, yet it may prove somewhat volatile and tough to call on a quarter-to-quarter basis once you're looking more than 1 or 2 quarters ahead.
Just a quick one about the Electron launches. Should we expect any ASP trends to hold back additional tailwinds during the second half? Or will things stay pretty stable over the coming couple of quarters?
My sense is that we'll keep seeing a steady shift, I'd say, as a greater share of our mix moves toward HASTE, which benefits the ASP. Margins, I think, stay fairly even, since although HASTE commands a higher price, it comes with considerably more mission assurance and other associated elements. So the dollar amounts are larger in absolute terms. Between HASTE and Electron, the gross margin percentage holds roughly the same. And then — so overall, I'd say we've witnessed quite a favorable rise in ASP across the past several years, driven by a bigger share coming from HASTE, and I don't expect that to change. As a matter of fact, we keep growing that subsegment of the business quite well. And once more, considering what Pete mentioned earlier about the Golden Dome and how vital hypersonics test capabilities are, that represents a really robust growth area for us moving forward; so on the whole, I think there's a positive lean toward a higher ASP per launch, exactly as we've observed over the past several years.
All right, and then perhaps a second question for Pete. Pete, could you maybe think back to how things were a few years back, when reaching that launch rate, the readiness of customer payloads was a variable thing? Has that shifted at all? Has the character of the customers shifted such that you feel more at ease about achieving a cadence of 11 to 13 days? Is it simply a larger count of customers arriving, so you can sort of unload them? Or whatever insight you can offer would be useful.
Yeah. Thanks, Suji. I'd simply say we've likely become better at resembling a duck, sitting on a glassy pond looking ordinary while its legs are going full speed underneath; and a greater cadence lets us relocate customers around. So I'd say that's simply how the Launch business works, payloads are ready until they aren't. I believe we've simply become far better at handling those customers, having additional rockets integrated, prepared to fly, and handling all of that. So it's wonderful to hear that it appears smooth, but off-camera, everybody is going full tilt, shuffling and swapping and ensuring everything appears smooth from the outside.
This is Justin Lang, filling in for Kristine. Pete, regarding the Neutron schedule, how do you avoid hitting the same problem with the Stage 1 tank? Would the program have achieved the prior target of reaching the pad in 1Q? Based on what you said before, it sounded like a substantial amount of qualification work got done during the quarter. So I'm just trying to figure out whether other elements are at play in this revised timeline, or whether it's truly confined to the Stage 1 tank problem?
Yes. Thank you, Justin; it's somewhat difficult to answer because when the tank failed, the reverberation traveled through the test stand and the whole company. So the instant that occurred, everyone halted their tasks, and it made a lot of sense to focus on the tank to determine the cause. So we shifted many resources from numerous areas of the business. So I would need to go back and examine whether, had we run everything forward, that timeline would have looked different. But it's rather hard right now because we had an anomaly.
One more point I'd tack on there. If anything good came out of that tank anomaly, it's this: what occurred gave the other subsystem groups a chance to run through all their demonstrations far more thoroughly than the tight timeline we'd been chasing would ever have allowed. So in a sense, the tank giving way means a test flight with notably less risk once it takes place afterward in 2024. So yes, I'd say — an anomaly never makes anyone glad; it's unplanned and surely not something we saw coming, yet I believe it does assist us in lowering the program's general risk profile as we head toward that initial test launch.
Understood, that clears things up and is useful. Also, Adam, I actually have just a single question for you regarding the SDA award. I'm wondering whether you might elaborate a bit further on the cash profile specifically and how it matches up with the revenue build curve you described previously?
Yeah. It's actually somewhat intriguing with programs like these, given how the accounting and rev-rec work, so under ASC 606, you — we do model these, but you must always remain in a positive cash position; so you — when you figure out your milestones and the manner in which you're sending dollars out to your subs and such and using funds within the program, you constantly need to be situated with positive cash so that you can recognize revenue over the course of it. And so this program fits with that. We've received some inquiries about whether the partial government shutdown has affected our customer, in this instance, the ability to pay, as they're aware; as a matter of fact, we received a very large payment from that customer. So the money continues to flow and everything appears to be green lights at this moment.
That wraps up our question-and-answer portion. With that, we're bringing today's program to a close, so you're free to hang up now. Thanks to all of you, and have a great remainder of your day.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete Rocket Lab Corporation transcript - plus 132,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.
Get an API key View API docs →For developers and AI pipelines
Programmatic access to Rocket Lab Corporation earnings transcripts and 132,000+ others is available through the
EarningsAPI REST API and the hosted MCP server.
Quarterly plans from $162 - full transcripts, speaker segments, full-text search,
and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.