The Trigger Fired: BlueBird 6 Is on Orbit and a $30M Prime Award Landed, Upgrading ASTS to Outperform

Results vs. Consensus

For six quarters the ASTS income statement was a lagging record of cash spent on a constellation that was not yet flying, and we treated the print as a secondary read behind the four forward variables (launch, funding, demand, spectrum). This quarter is the first where the print and the operating story point the same way. FY2025 revenue of $70.9M (against $4.4M in all of 2024) landed at the top of management's $50-75M guide, carried by a $54.3M fourth quarter that beat the Street's ~$40.7M line by roughly a third. The GAAP loss is still large and still mostly below-the-line, so the scorecard remains a partial read, but for the first time the revenue line is doing real work rather than rounding to zero.

Q4 2025 Scorecard

Full-Year 2025 Scorecard

Quality of the Beat/Miss

• Revenue: the beat is real and it is broad-based within the two live vectors. The $54.3M Q4 was gateway hardware sales (the leading indicator of MNO network readiness), U.S. government service-milestone achievements, and a new item, a critical-consulting engagement for an MNO partner. None of it is commercial-service revenue yet, so this is a build-phase beat: partners paying for equipment and the government paying for milestones. The FY $70.9M at the top of a range management set six months ago is the more important fact than the Q4 optics, because it is the first guide ASTS has met at the top rather than the low end.
• EPS: the $(0.26) Q4 loss to common (and the $(1.34) full year) is a poor proxy for the operating trajectory, as it has been every quarter. Operating expense drove only part of it; the majority of the loss sits below the line in interest, financing costs, and non-cash fair-value marks on the warrant and derivative liabilities tied to an enlarged convertible and debt stack, with a meaningful slice of the pre-NCI loss allocated to the noncontrolling interest under the Up-C structure. For a company whose equity is a call option on a constellation, reported EPS will stay disconnected from value until service scales.
• Opex discipline (the watch item, resolved): after two straight quarters printing above guide, adjusted opex ex-COGS of $66.8M landed squarely in the mid-$60s management had guided. That ends the overrun streak we flagged at Q3 and restores some credibility to the run-rate. The headline $95.7M adjusted figure is up $28.0M sequentially, but $23.4M of that is new adjusted cost of revenue tied to gateway deliveries, i.e., the cost of actually recognizing revenue, not runaway spend.
• Capex (the new watch item): Q4 capex of ~$406.7M ran well above the $275-325M guide, and full-year capex reached ~$1.065B. Management framed the overage as intentional, pulling forward satellite-material purchases and launch-contract payments to protect the 2026 cadence. That is a defensible reason, but a ~$82M overshoot on a figure guided just one quarter earlier is worth tracking, especially with Q1 2026 capex now guided to a still-higher $350-425M.

Grading Last Quarter's Commitments

At Q3 we recorded a nine-point checklist of what management committed to on that call. A recap is a chapter in a running story, so we grade those commitments against this quarter's actuals before assessing anything new.

Read: five of nine met or exceeded, including the two that gated the rating (a Block 2 on orbit and a signed, quantified government award). The misses cluster on cadence and near-term service timing (launches-by-Q1 and nationwide service both slipped), and capex ran hot. The scorecard is the upgrade in miniature: the high-order commitments were delivered, the timing commitments continue to slip, and on balance the delivered ones matter more.

Business-Line Performance

ASTS still does not report conventional P&L segments; revenue flows through three vectors (government, gateway hardware, and now a nascent MNO-services line) with commercial subscriber service still pre-revenue. We assess those vectors alongside the operating and capital metrics that govern the cash runway, and the constellation KPIs that are, for a pre-service builder, the real scorecard.

The Launch: BlueBird 6 on Orbit

The defining event of the quarter, and of the coverage to date, is that BlueBird 6 launched and completed its on-orbit unfolding. This is the first Block 2 BlueBird, an array of roughly 2,400 square feet, about 3.5x larger and 10x the capacity of the five Block 1 BlueBirds already flying, and management described it as the largest commercial communications array ever deployed in low earth orbit. Unfolding a structure of that size in space is the step every skeptic (this one included) flagged as the binary risk, and it worked.

"BB 6, it is the largest phase array ever deployed in space. It's 3.5x bigger than our previous deployments... going through that first deployment of 2,400 square feet successfully, learn how to capture, control and manage the satellite at that size will allow us to actually do it much more faster." — Abel Avellan, Chairman & CEO

Assessment: This is the fact that changes the rating. A satellite on a pad in India earned nothing and proved nothing, as we wrote at Q3; a satellite that has unfolded a record-size array on orbit proves the single most important thing the company had never demonstrated. It does not prove the cadence, and it does not prove that 45 more can follow this year. But it converts the largest technical unknown in the story from "unproven" to "demonstrated," which is precisely the threshold we said would move us.

U.S. Government: A Hard Number, Finally

The government leg produced the second half of the trigger. AST SpaceMobile USA, the wholly owned defense subsidiary, was selected as a prime contractor on the Space Development Agency's HALO Europa Track 2 program, a $30M award focused on resilient, low-latency tactical direct-to-device communications. Separately, the company received an IDIQ award under the Missile Defense Agency's SHIELD program and continues to execute against 10 existing government contracts and the Golden Dome missile-defense initiative. Management stressed that government revenue scales by satellite count rather than depending on the full constellation, making it an early, reliable contributor.

"We also recently announced our status as a prime contractor to the U.S. government and received a $30 million contract award from the United States Space Development Agency for the Europa Track 2 commercial solutions program... The award further validates the dual-use nature of our technology." — Scott Wisniewski, President

Assessment: At Q3 we graded the unquantified, shutdown-contingent prime "award" as progress, not resolution, and explicitly said it did not clear the "signed material award" bar. A $30M prime contract, publicly announced with a program name and a dollar figure, clears it. It is not large relative to the company's spend, but it is signed, it is quantified, and prime-contractor status positions ASTS to compete for the follow-on programs of record management frames as billions in aggregate potential. This is the government leg converting from narrative to contract.

Commercial: The Book Grows to $1.2B and the Gateways Ship

The commercial story deepened without a single new headline dominating it. Aggregate contracted revenue commitments rose to over $1.2 billion (from the $1B+ first disclosed at Q3), the definitive Verizon and stc agreements are now in place, and stc's $175M prepayment was received in 2025, closing a commitment we were watching. Fifteen commercial gateways were delivered in H2 2025 across nine customers on five continents, and management announced expanded initiatives with Orange, Telefónica, CK Hutchison, Taiwan Mobile, and Sunrise, plus the formal unveiling of Satellite Connect Europe, the Vodafone distribution JV.

"We delivered 15 commercial gateways to MNO partners in the second half of 2025... a well-diversified set of initial gateway deliveries across nine different customers across five continents, which starts to paint the picture of our initial commercial markets." — Scott Wisniewski, President

Assessment: The gateway deliveries are the tell. An MNO does not buy and install ground infrastructure across five continents unless it is preparing to route real traffic; fifteen units to nine customers is a physical, hard-to-fake signal of partner conviction ahead of service. The $175M stc prepayment landing on schedule is a second such signal, and the backlog stepping to $1.2B keeps the demand pillar on track. None of it is subscriber revenue yet, but the commercial side is doing exactly what it should ahead of a 2H 2026 service start.

Operating & Capital Metrics

Assessment: The line that read zero for six quarters now reads one. That single change reframes every metric above it: the capex is buying satellites that can now be shown to work, the opex is funding a factory whose output has reached orbit, and the revenue is the earliest trickle of a model that turns on with the constellation. The capex overrun is the one genuine negative in the table, and it is a timing-and-acceleration story rather than a cost-inflation one, but it bears watching against the higher Q1 guide.

Constellation & Launch Milestones (the KPIs that actually matter)

Assessment: The table tells a more balanced story than the headline. The top row flipped from zero to one, which is the whole re-rate. But the third row, launches by end Q1, is a clear miss against the target management set last quarter: one satellite is up and BB7 is still on the ground. The path from one on orbit to 45 by year-end at a one-to-two-month cadence, on stacked launches that have not yet flown, is the 2026 crux, and it is exactly as unproven at scale as it was demonstrated at unit-one. We upgrade on the demonstrated unit; we stay honest about the unproven scale.

Key Topics & Management Commentary

Overall Management Tone: Management was the most assured it has been across our coverage, and for the first time the confidence rested on a completed milestone rather than a forward promise: the call opened on BlueBird 6 being on orbit rather than on a launch manifest. The posture on funding and demand was matter-of-fact to the point of routine, and the government commentary carried a new hard number. Where management stayed conspicuously general was the go-forward launch cadence, deflecting specific launch-date questions to "we're not going to speculate on launch timing" while pointing to production progress, the same substitution of a factory answer for a launch answer we flagged last quarter.

BlueBird 6 and What the Deployment Proved

Management treated the BB6 unfolding as both a milestone and a learning event, arguing that having controlled a 2,400-square-foot array once, the next deployments (in stacks of three, four, six, or eight) get faster. The framing matters because the entire 2026 plan rests on repeating this step dozens of times.

"That was a very, very important milestone in learning how to operate, deploy and fly something of this size, which will help us to do it faster in the next deployments. The other thing that will happen going forward passing 6 and 7 is that we're stacking the satellites." — Abel Avellan, Chairman & CEO

Assessment: The "we learned to fly it, now we go faster" logic is credible and is the right thing to have proven first. But there is an unavoidable gap between deploying one satellite and deploying stacks of six or eight on a vehicle configuration that has not yet flown stacked. The milestone de-risks the array; it does not yet de-risk the cadence.

Launch Cadence and the Manifest

The go-forward manifest is BB7 on a New Glenn in March, then launches every one to two months on average, with 12 additional contracted launches across several vehicles and a recently signed standby agreement with a new heavy launch vehicle. Management leaned on the New Glenn 7-meter fairing (up to eight satellites) and, importantly, on this being the first New Glenn flight to reuse a previously flown first stage, which it argued underpins a 30-day-or-less booster turnaround.

"We continue to expect launches planned every one to two months on average, starting with our first New Glenn launch expected in March... this launch will be the first New Glenn launch to use a previously flown first stage... we expect the New Glenn boosted to be reused every 30 days or less after our upcoming launch." — Abel Avellan, Chairman & CEO

Assessment: The cadence case now rests heavily on New Glenn, a vehicle early in its own operational life, hitting a reuse tempo it has not yet demonstrated. The "five launches by end Q1" target from last quarter has quietly lapsed to one. We credit the contracted manifest and the stacking capability, but the every-one-to-two-months cadence is the single biggest thing that still has to be proven, and it depends on third-party vehicles as much as on Midland's factory.

Manufacturing Ramp and Stacking

The factory continues to be the more grounded half of the equation: 29 Block 2 satellites in various stages of production, on track for 40 satellite-equivalents of microns in 1H 2026, at a six-per-month cadence, on ~95% vertical integration across soon-to-be over 0.5M square feet (a fourth Midland site was acquired for dedicated micron production). The new operational wrinkle is stacking, engineering satellites to launch in groups rather than singly, which management described as near-complete.

"After Blue Bird 7, our satellite will support a stackable configuration of three, four, six, or eight satellite per launch, which allow us to meet our 2026 deployment goals." — Abel Avellan, Chairman & CEO

Assessment: Twenty-nine in production versus 19 last quarter is tangible, and the stacking capability is the mechanism that makes 45 launches in a year arithmetically possible. But finished inventory in Midland is still not on orbit, and the bottleneck was never the microns; it is the launch slots. The production line is doing its job. The manifest has to do the rest.

Unit Economics and the Cost-Per-Satellite Anchor

Management reiterated a cost of $21-23M per satellite (direct materials plus launch) for a constellation of 90-plus Block 2 BlueBirds, with the caveat that geopolitical factors could move launch costs. The ~95% vertical-integration strategy, extended this quarter with a fourth Midland site dedicated to micron production and an expanded Homestead, Florida footprint, is the lever management credits for keeping component costs low and controlling the process end-to-end while securing long-lead materials in advance.

"We continue to estimate that the average capital cost, including direct materials and launch costs for our constellation of over 90 Block 2 BlueBird satellites will fall in the range of $21 million to $23 million per satellite." — Andy Johnson, CFO & Chief Legal Officer

Assessment: A stable $21-23M per-satellite anchor is useful because it lets an investor size the remaining build: roughly 90-plus satellites at ~$22M implies a program cost the ~$3.9B balance sheet comfortably covers, which is the arithmetic behind "fully funded to 100-plus." The vertical-integration edge is real and is the reason the cost has held flat across quarters. The open variable embedded in the range is launch cost, which loops back once more to the third-party vehicle dependency that governs the whole 2026 plan.

The Balance Sheet: Funded Past the Full Constellation

Pro-forma cash and liquidity reached ~$3.9B (from ~$3.2B at Q3) after a February 2026 convertible offering (2.25% coupon, ten-year, ~$116.30 effective strike, over $1B raised) and ~$706M of Q4 ATM proceeds, against ~$2.8B of GAAP cash and ~$2.2B of long-term debt. Management also equitized $457M of the $460M January converts and $250M of the $575M July converts into Class A shares, and reiterated no plans for further convertible debt.

"We are now not only fully funded to manufacture and launch a constellation of over 100 satellites to provide worldwide space mobile service, but we have increased our financial flexibility... At this time, we do not have any plans to pursue additional convertible debt." — Andy Johnson, CFO & Chief Legal Officer

Assessment: The funding debate is not merely won, it is over. Raising at a 2.25% ten-year coupon with a strike well above the current price, above the first-100-satellite need, buys spectrum-deployment speed, AI-monetization optionality, and higher-cost-debt reduction. The cost is more shares and heavier below-the-line drag on reported EPS, which is the correct trade for a call-option-on-a-constellation equity. This remains the strongest pillar in the thesis.

The Path to ~$1B of Revenue in 2027

Management set out a three-year revenue arc for the first time: at least double 2025 in 2026 (to a guided $150-200M), and a 2027 goal approaching ~$1 billion, weighted toward commercial as service scales, with government multiplying alongside. 2027 would be the first full year of commercial-service revenue, and management characterized the ~$1B as mostly long-term contracted or recurring.

"We see the opportunity in 2027 approaching $1 billion in annual revenue, importantly comprised of revenue both long-term contracted or highly recurring in nature, subject to achievement of commercial and government service objectives." — Scott Wisniewski, President

Assessment: The 2027 number is a goal, not a guide, and it is heavily conditioned on the constellation reaching service scale, which loops back to the cadence question. But the structure of the target (contracted plus recurring) and the doubling in 2026 that precedes it give the terminal case a shape it lacked. We treat the $150-200M 2026 guide as the accountable number and the ~$1B as the option value.

Services Economics: ~90% Gross Margins

A new disclosure this quarter: services gross margins are running around 90%, and management argued steady-state EBITDA margins could reach the 90% area or higher, citing a fixed cost base, minimal true variable cost, and a revenue-share go-to-market that is "not just wholesale, but super wholesale."

"This has just tremendous operating leverage in it... at this point, our flow-through margins and our operating leverage, we think, over time, could contribute to an EBITDA margin in the 90% area or higher." — Scott Wisniewski, President

Assessment: Ninety-percent-plus flow-through is consistent with the best of the satellite industry and with the fixed-cost, capacity-sale nature of the model. It is the mathematical reason a company burning nine figures a quarter can become highly cash-generative once the constellation is up. The margin structure was always assumed; disclosing ~90% services gross margin puts a number on the terminal profitability case.

Spectrum and the Mid-Band Constellation

The spectrum position is broadly unchanged but gained a timeline: management now plans to begin launching a mid-band constellation (combining operator 3GPP spectrum with its own L- and S-band) by the end of 2026, which it argues lifts data rates well above the current 120 Mbps by aggregating over 100 MHz in certain regions. Access spans ~1,150 MHz of tunable low- and mid-band, with L-Band still awaiting FCC approval and spectrum costs still capitalized (not yet in opex) until monetization begins.

"We're planning to start launching the mid-band constellation by the end of the year... which combined give a great flexibility to the offering and also allow us to continue to increase the data rate capacity... way above our 120 megabit per second that we already have in Block 1." — Abel Avellan, Chairman & CEO

Assessment: The multi-band, software-tunable position remains the least-replicable moat asset ASTS owns, and putting a launch timeline on mid-band gives it a nearer edge. FCC approval on L-Band is still the gating regulatory event, and the spectrum-cost line will eventually move into opex once monetization starts, a future modeling item. For now the pillar is on track.

Adjacent TAMs: Radar, AI, and Spectrum Multiplication

Management spent unusual airtime, prompted by shareholder and analyst questions, on what the large-aperture, high-power platform can do beyond cellular broadband. The list ran to radar, precise geolocation, on-orbit data management, and an "AI spectrum management" concept in which combining the large aperture with AI multiplies effective spectrum, described as potentially 3x to 10x the 50 MHz baseline. The February $1B raise was explicitly tied in part to monetizing AI-related opportunities "increasingly coming our way."

"We believe in combining our large aperture with our AI capability will create a multiplier for the spectrum. So the 50 megahertz that we have, it will feel a multiple of that. It could be 3x that, it could be 10x that... including very precise geolocation, radar, communications, all that wrap up with an AI infrastructure." — Abel Avellan, Chairman & CEO

Assessment: These adjacencies are real capabilities of the platform and genuine optionality, and the dual-use radar/geolocation angle dovetails with the defense leg that just produced a signed prime. But they are, for now, TAM-expansion narrative rather than modelable revenue, and the "AI opportunities" framing attached to a fresh $1B raise warrants the same show-me discipline we apply to everything else here. We treat these as free options embedded in a constellation that must first fly at scale, not as a reason to pay more today.

Guidance & Outlook

Guidance is best read, as always with ASTS, as a set of operational commitments rather than a P&L model. The accountable new number is the FY2026 revenue guide of $150-200M, at least double 2025, which management said is derisked by the contracted pipeline (roughly half already booked or contracted) with upside from initial commercial service in 2H 2026 and additional government wins. Management explicitly declined to guide quarterly, arguing the business is best judged annually given lumpy milestone and equipment timing, and cautioned against using the outsized Q4 as a run-rate.

Implied 2026 shape: doubling FY2025's $70.9M implies ~$142M at the low bound of "at least double," and the $150-200M guide sits just above that. The build is gateway deliveries plus government milestones plus MNO consulting in the first half, with commercial-service revenue only beginning in 2H. Management pointedly said not to build quarter-over-quarter off the $54.3M Q4, so a softer Q1 (analyst feeds near ~$32M) followed by a second-half ramp is the expected cadence.

Street at: for a name at this stage, the meaningful benchmarks are management's own $150-200M 2026 guide and, above all, the launch cadence. Consensus revenue estimates exist but are dispersed and of limited use; the FY2026 print will be judged against the guide, and the stock against the pace of satellites reaching orbit.

Guidance style: management continues to set ambitious operational timelines and to let the specific ones (five launches by Q1, nationwide service) slip while holding the annual and terminal targets. The FY revenue guide has now been met at the top once, which earns it more credibility than the launch-date guidance, which keeps slipping. We read the 2026 revenue guide as achievable and the cadence guidance as carrying continued downside skew on timing.

Analyst Q&A Highlights

Manufacturing Pace and Whether the Bottleneck Has Cleared

The most thesis-relevant exchange put the quiet part on the table: the manufacturing pace is behind where management had expected it to be, and an analyst asked directly what had taken longer and whether it was now resolved. Management pointed to the microns being on satellite 30 and to the stacking process being near-complete, framing the delay as the one-time difficulty of scaling to a 3.5x-larger satellite and then learning to stack them.

Q: "I think clearly, the manufacturing pace is somewhat behind where you had expected it to be. Perhaps you could maybe just give us some appreciation for the kinds of things that maybe took longer than you had expected and whether you think you've now worked through all those issues?"
— Bryan Kraft, Deutsche Bank

A: "I think we are at a point where you see that acceleration. We certainly see that in the manufacturing of the key building block, which is the Micron... we are on satellite 30... what is something that helps us to accelerate our cadence of satellites in orbit is we are able to stack them. And that stack is difficult... And that is near completion."
— Abel Avellan, Chairman & CEO

Assessment: The candor that the pace slipped is welcome, and the explanation (bigger satellite, then the engineering of stacking) is coherent. But the answer again addresses the factory rather than the launch manifest, which is where the cadence risk actually lives. Production reaching six per month does not create launch slots. This is the crux the 2026 story turns on, and it is only half-answered.

Timing of the Next Launch After BlueBird 7

An analyst pressed on whether a multi-satellite launch could come as soon as April or whether it would be two months behind the March BB7 launch. Management confirmed all launches after BB7 are stacked (no more singles), said the next batch is expected to ship in April, and declined to commit to a launch date beyond the roughly three-week ship-to-launch window under ideal conditions.

Q: "I know you said that BB 7 is expected to go up this month and then launches every one to two months. Could we expect possibly a launch with multiple satellites in April? Or is it likely to be two months post the March launch?"
— Bryan Kraft, Deutsche Bank

A: "We expect to ship that next batch in April. So depending on timing and of course, under ideal conditions, it's about three weeks or so to launch from there. So we're not going to speculate on launch timing for that, but... we look like we're going to be in a position to ship those in April."
— Scott Wisniewski, President

Assessment: "Ship in April, we won't speculate on launch timing" is the honest but telling answer. Management will commit to a factory milestone (ship date) but not a launch date, which is the same asymmetry we have seen all along. The first stacked launch is the real proof of cadence, and its timing remains uncommitted.

How Much of the 2027 ~$1B Is Already Contracted

An analyst sought to pin down how much of the ~$1B 2027 goal is contractually locked via minimum commitments versus dependent on usage-based subscriber uptake. Management said the $1.2B backlog, while a proud milestone, contributes only a minority to any single year, framing per-year contracted contribution in the low hundreds of millions.

Q: "For the $1 billion revenue goal for 2027, how much of that is customer or subscriber usage based versus being like minimum revenue commitments that are contractually obligated... how much of that $1 billion is then like already in the bag?"
— Louie DiPalma, William Blair

A: "We're at $1.2 billion contracted backlog right now... but in terms of its contribution to each individual year, it will be a minority for sure. So if we're... in terms of a goal of $1 billion, you can think of that in the low hundreds of millions, somewhere in $100 million to $300 million range depending on the year."
— Scott Wisniewski, President

Assessment: A useful and honest calibration. Roughly $100-300M of any year's revenue is contracted; the balance of the ~$1B 2027 goal is usage-based and therefore dependent on subscriber uptake and, upstream of that, on the constellation reaching service scale. It reinforces that the backlog is a floor and a financing tool, not the whole picture, and that the terminal case still leans on execution.

Services Gross Margins and Steady-State Operating Leverage

An analyst asked whether the newly disclosed ~90% services gross margin is a fair long-term proxy and where steady-state EBITDA margins could land. Management confirmed the ~90% flow-through and pointed to a fixed cost base, minimal variable cost, and the revenue-share model as the drivers of 90%-plus EBITDA over time.

Q: "Looking at your new disclosure, it appears your services gross margins are around 90%. Is this a good way to think about the business longer term? ... where you believe steady-state EBITDA margins can reach for the business?"
— Chris Schoell, UBS

A: "We've been pretty consistent about this over time... this has just tremendous operating leverage in it, and we've always known that off of a fixed cost base... our flow-through margins and our operating leverage, we think, over time, could contribute to an EBITDA margin in the 90% area or higher."
— Scott Wisniewski, President

Assessment: The margin structure is the crux of the terminal-value case, and putting ~90% on the services gross margin makes the "why own this at scale" math explicit. The caveat is the usual one: 90% flow-through on revenue that does not yet exist. The disclosure sharpens the reward side of the risk/reward, which is part of why the upgrade is justified now.

Why Raise Another $1B in February After Declaring Full Funding

A pre-submitted shareholder question went at the apparent redundancy of the February convertible when the company was already funded to 100-plus satellites. The CFO framed it as pure flexibility for opportunities beyond the first 100: accelerating spectrum deployment, monetizing technology for AI use cases, government investment, and higher-cost-debt reduction, while reiterating no further converts are planned.

Q: "Can you share more color on the most recent $1 billion convertible note offering? Many investors are confused as your current liquidity was already approximately $3 billion... Were there any specific opportunities in mind... or is it really just in case something pops up?"
— Kevin (shareholder), submitted in advance

A: "In Q4, when we finished the convertible in October, we were in a position to fully fund the worldwide constellation at 100-plus satellites... The convertible deal that we did at just over $1 billion in February provides us essentially extra flexibility to look at investments that go beyond that first 100 constellation... And I would just close by noting that we've confirmed that we have no current plans to look at an additional convertible deal."
— Andy Johnson, CFO & Chief Legal Officer

Assessment: A clean answer. Raising at a 2.25% ten-year coupon above the funded need, with the stock well below the strike, is defensible opportunistic capital allocation, and the explicit "no further converts planned" caps the dilution narrative. The mild concern is that "flexibility for things beyond the first 100" is open-ended; disciplined deployment of that cash is now something to watch.

Dual-Use Platform and Whether Government Needs Dedicated Satellites

An analyst asked whether future satellites might carry government-exclusive payloads. Management was firm that a single platform serves both the commercial and government TAMs simultaneously, and that no dedicated government constellation is needed, which is the economic heart of the dual-use model.

Q: "Do you see any scenario where you build and launch future BlueBird satellites with different payloads that might be exclusively for government customers or applications?"
— Griffin Boss, B. Riley Securities

A: "The satellites are really designed to manage all these applications in a single platform. So we do not need multiple satellites for multiple payloads... we want to maximize and take advantage of a platform that can be used simultaneously for the two TAMs."
— Abel Avellan, Chairman & CEO

Assessment: The single-platform answer is the right one economically: one constellation serving both the $1.2B commercial book and the government programs of record without duplicating capital. The flip side, which we noted at Q3 and remains true, is that government revenue therefore scales with the same launch cadence as everything else, so it is not an independent hedge against launch risk, even with the $30M prime now signed.

Whether Q4 Is a Baseline for 2026

An analyst asked whether the $54.3M Q4 should be treated as a run-rate baseline for 2026 or whether growth should be measured off the annual number. Management was explicit: think annually, not quarter-over-quarter, because pre-service revenue is lumpy and the commercial line only turns on in the second half.

Q: "Is it fair to use the 4Q performance as a baseline for 2026 and then growing from there? Or is the commentary in terms of growth for '26 more aligned with just growing from the 2025 annual number?"
— Colin Canfield, Cantor Fitzgerald

A: "Quarter-to-quarter, I wouldn't say we're planning on building quarter-to-quarter. I think about it annually... at least doubling where we hit in 2025 is the right way to think about it with, of course, upside as we launch commercial service. But quarter-to-quarter, at least in the next few quarters before commercial service comes into play in the second half of 2026..."
— Scott Wisniewski, President

Assessment: A necessary expectation-setter. The $54.3M Q4 was inflated by lumpy gateway and milestone timing, and management is right to steer off it. The practical implication is that a soft Q1 print should not be read as a stumble; the accountable measure is the $150-200M full year, and the stock-moving measure is satellites on orbit.

Thermal-Management IP and Data-Center-in-Space Optionality

An analyst probed a specific patent covering localized thermal management at each antenna assembly, asking how it and the broader IP portfolio might enable adjacent opportunities such as on-orbit compute for AI. Management used the question to walk through the enabling technologies it had to invent (low-cost power generation at scale, dissipating high wattage per square meter within space constraints) and argued these underpin both the core broadband product and the adjacent radar/AI/data-management use cases.

Q: "You have this one specific patent... for thermal management systems for structure in space... So just curious if you could maybe elaborate on that specifically as well as your other capabilities and how that could potentially be used for opportunities in data centers in space?"
— Griffin Boss, B. Riley Securities

A: "There are many key enablers that needed to be designed by us... the ability to generate power at a low cost per square meter and then being able to dissipate and effectively run a lot of wattage per square meter within the power constraints of space... That's why we have built up a significant portfolio of IP... it opened a lot of other opportunities on the TAM that we have."
— Abel Avellan, Chairman & CEO

Assessment: The power-and-thermal engineering is genuinely the hard part of operating a record-size array, and it is credible that the same IP underpins adjacent use cases. But the data-center-in-space framing is speculative optionality, not a near-term revenue line, and management was careful to keep the emphasis on broadband as the largest TAM. We file this under free options, useful for the terminal narrative, immaterial to the next twelve months.

What They're NOT Saying

With the two gating milestones delivered, the omissions this quarter cluster around timing and specifics rather than around the existential questions that dominated prior calls. That is itself progress, but the gaps below are where an investor's remaining doubt should concentrate.

1. A committed launch date beyond BB7: the single most important go-forward number. Management will commit to shipping the next batch in April but explicitly "won't speculate on launch timing," and the "five launches by end Q1" target from last quarter simply went unmentioned rather than being reconciled. The first stacked launch is the real cadence proof, and its date is undisclosed.
2. Why the capex overran by ~$82M: Q4 capex of ~$407M ran well above the $275-325M guided just one quarter earlier. "Intentional acceleration and launch-payment timing" is offered, but a nearly 25% overshoot on a one-quarter-old guide, with Q1 guided higher still, gets a one-line explanation rather than a bridge.
3. Quarterly 2026 revenue guidance: management gave a full-year $150-200M range and then declined to shape it quarterly, which is defensible for a lumpy pre-service business but also conveniently removes the near-term accountability checkpoints between now and the 2H service start.
4. What "nationwide intermittent service" became: last quarter's "early 2026" nationwide intermittent U.S. service was not reaffirmed with a date; the framing shifted to beta offerings "in the summer" and commercial activation in 2H 2026. The near-term service milestone quietly moved right without being called out.
5. The specifics of the "beyond the first 100" investments: the February $1B raise is earmarked for spectrum acceleration, AI monetization, government, and debt reduction, but none of these carries a size or timeline. Open-ended optionality funded by fresh convertible debt is worth a closer eye on deployment discipline.
6. L-Band FCC approval progress: the mid-band constellation is now slated to begin launching by end-2026 and depends on L-Band, yet the FCC approval status, still pending, got no update. For a 2026 event gating a whole spectrum layer, the silence is notable.
7. BlueBird 6's operational performance data: management confirmed BB6 unfolded and expressed happiness with what it is seeing, but disclosed no measured throughput, coverage, or link-budget figures from the on-orbit satellite. For the milestone the entire re-rate hinges on, hard performance numbers, rather than "we're extremely happy," would have been the most valuable disclosure of the call.

Market Reaction

• Pre-print setup: ASTS closed at $86.92 on March 2 entering the print, up 19.7% year-to-date but down 21.8% over the trailing 30 days (from $111.21 on January 30) and well off its 52-week closing high of $122.09. Over the trailing twelve months the stock was still up 220.6%. The S&P 500 was +0.5% YTD. The setup was a name that had run enormously over the prior year but had already surrendered roughly a fifth of its value in the month into the print.
• After-hours / next-day session (March 3): shares opened at $88.01 (+1.3% gap), traded a wide $84.01 to $97.57 range (-3.3% to +12.3% intraday), and closed at $92.68, up 6.6% (+$5.76) on volume of 21.8M shares versus a 15.3M 30-day average (1.4x). The S&P 500 fell 0.9% the same day, so the move was roughly +7.5% on a relative basis.

The reaction is the mirror image of last quarter. At Q3 a headline-dense print on commitments and funding drew a muted, slightly negative tape because the market wanted a satellite on orbit, not more paper. This quarter delivered the satellite, and the stock rallied +6.6% into a down market on 1.4x volume off a beaten-down 30-day setup. The wide intraday range (a 15-point span from $84 to nearly $98) says the debate is live, but the close near the highs and the above-average volume say the marginal buyer showed up for the milestone. The tape did exactly what its Q3 silence implied it would: it moved on the launch, not on the commitments.

The relative move matters. On a day the S&P fell 0.9%, ASTS closing +6.6% is a ~7.5-point relative outperformance, and it came on a stock that had de-rated ~22% over the prior month, i.e., into a setup with far less froth than the ~$122 January high. That combination, a demonstrated catalyst against a de-risked entry, is a large part of why we are comfortable upgrading here rather than chasing at the high.

One note of caution on reading too much into a single session: the $84-to-$98 intraday range is characteristic of a high-beta, heavily-shorted retail favorite, and days like this can round-trip. We do not anchor the upgrade to the reaction-day tape; we anchor it to the milestone the tape was reacting to. The price action is confirmation, not thesis.

Street Perspective

The post-print debate reorganized around a new center of gravity. For two quarters the Street argument was whether commitments and funding were enough; now that a satellite is on orbit and a government prime is signed, the debate has moved to whether one satellite justifies a re-rate and whether the cadence can scale. We frame the five live debates below.

Debate: Does BlueBird 6 On Orbit Justify a Re-Rate?

Bull view: the single biggest unknown in the entire business, whether a record-size phased array could be deployed on orbit, is now demonstrated; the technical risk that justified a discount is retired, and the stock should re-rate toward execution-in-progress rather than execution-unproven.

Bear view: one satellite is not a constellation; the "five launches by Q1" target already lapsed, BB7 is still on the ground, and the every-one-to-two-months cadence on stacked launches has not flown once, so nothing about the 2026 deployment risk has actually been resolved.

Our take: both are right, which is why this is a measured Outperform, not a high-conviction one. The deployment risk on the array is genuinely retired, and that is the pre-committed trigger we said would move us. The cadence risk at scale is genuinely unresolved. The upgrade reflects that the former is the higher-order unknown, and demonstrating it is worth more than the schedule slip costs.

Debate: Is the $30M SDA Award Meaningful or Symbolic?

Bull view: it is the first signed, quantified U.S. government prime contract for the company, it validates the dual-use architecture, and prime-contractor status is the gateway to programs of record management frames as billions in aggregate; the dollar figure understates the strategic significance.

Bear view: $30M is immaterial against a company spending over $400M of capex in a single quarter; it is a foot in the door, not a revenue engine, and government procurement is slow, milestone-gated, and scales with the same launch cadence as everything else.

Our take: we lean bull, as we have on the government leg since initiation. The dollar size is small, but the bar we set was "signed and quantified material award," and a publicly announced prime contract with a program name clears it where last quarter's unquantified, contingent "award" did not. The value is the prime status and the optionality on follow-ons, not the $30M itself.

Debate: Is the Valuation Reasonable After the Pullback?

Bull view: the ~22% drawdown from the January high into the print removed froth, the balance sheet is funded past the full constellation, revenue hit the top of guide, and now the core technical risk is demonstrated; a de-rated entry into a de-risked story is an improved risk/reward.

Bear view: even at ~$93 the stock is up 220% over twelve months and carries a large-cap valuation on a company with $70.9M of annual revenue and no commercial service; it still discounts the constellation succeeding, so a lower price on a still-binary outcome is a better entry, not a margin of safety.

Our take: the valuation is still full on any near-term metric, and ASTS remains a call option on a constellation. But the option is worth more now that the array-deployment risk is retired and the government leg is contracted, and the entry is ~22% better than a month ago. We are not paying up at the highs; we are upgrading on a de-rated price after the binary began to resolve in the bulls' favor. That is the difference between this quarter and the last two.

Debate: Will the 2026 Launch Cadence Materialize?

Bull view: the manifest is contracted (12 additional launches plus a new standby vehicle), stacking is near-complete, New Glenn's larger fairing carries up to eight satellites, and booster reuse should unlock a 30-day turnaround; the pieces for 45 in orbit by year-end are in place.

Bear view: the cadence now rests on New Glenn hitting a reuse tempo it has never demonstrated, the "five by Q1" target already slipped to one, and every prior ASTS launch timeline has slipped; 45 satellites in orbit in ten months from a base of one is a very steep ramp.

Our take: this is the live risk, and it is why the upgrade is measured. We credit the contracted manifest and the stacking capability, but the every-one-to-two-months cadence is unproven and vehicle-dependent. We would treat a genuine multi-satellite stacked launch in Q2, and a second batch behind it, as the confirmation that the cadence is real. Until then, the 45-by-year-end target carries real downside skew.

Debate: Does the Moat Hold Against Other Direct-to-Device Entrants?

Bull view: ASTS is architecturally differentiated toward genuine broadband (large arrays, the AST-5000 ASIC, >120 Mbps) rather than emergency texting, the multi-band owned-plus-partner spectrum position is hard to replicate, 3,100-plus patent claims raise the barrier, and now it has actually deployed the record-size array that competitors have only described.

Bear view: better-capitalized rivals with their own launch capacity and existing constellations can iterate quickly, carriers may commoditize basic satellite-to-phone connectivity, and being first to a broadband spec matters less if a rival reaches "good enough" scale first.

Our take: the broadband-versus-texting distinction is real and, as of this quarter, partly demonstrated rather than purely on paper, because BB6 is the array the differentiation depends on. The spectrum-plus-patent position remains the durable moat. Competitors still have to deploy comparable capacity; ASTS just did it first at Block 2 scale, which strengthens the moat argument relative to last quarter.

Capital Structure & Dilution Mechanics

Because the reported loss and the share count both keep moving on financing rather than operations, a brief reconciliation helps readers square the $(0.26) Q4 (and $(1.34) full-year) loss with the operating story.

• The below-the-line drag. As in prior quarters, a large share of the net loss sits below operating income in interest, financing costs, and non-cash fair-value marks on warrant and derivative liabilities tied to the convertible and debt stack. These marks swing quarter to quarter (the Q4 loss to common of $(74.0)M is actually narrower than Q3's $(122.9)M, not because operations improved but because the marks moved the other way), which is why reported EPS is a poor read on the trajectory.
• The de-levering trade. Management equitized $457M of the $460M January 2025 converts into 19.2M Class A shares and $250M of the $575M July 2025 converts into 4.5M Class A shares. This adds shares but removes near-strike, higher-coupon convertible debt and its refinancing risk, replacing it with a longer-dated, lower-coupon (2.25%, ten-year) structure struck at ~$116.30, above the current price.
• The funding cushion. Pro-forma liquidity of ~$3.9B against ~$2.2B of long-term debt and ~$1.6B of capitalized property and equipment leaves the company funded past the 100-plus-satellite constellation, with the February raise explicitly earmarked for opportunities beyond it.

Assessment: The net effect is a cleaner, longer-dated liability stack funded past the full constellation, at the cost of more shares and a heavier, lumpier below-the-line non-cash drag on reported EPS. For a call-option-on-a-constellation equity, that is the correct trade; the reported loss will remain a poor proxy for value until service scales, and readers should judge the company on cash, capex, satellites on orbit, and revenue against guide rather than on GAAP EPS.

Model Update Needed

Valuation impact: ASTS is not valuable on any near-term cash-flow multiple; it remains a call option on a constellation, and the near-term GAAP loss is a poor proxy for value. What changed this quarter is the probability weighting on the option: the array-deployment risk is retired (BB6 on orbit), the government leg is contracted ($30M SDA prime), the balance sheet funds the whole build (~$3.9B), revenue hit the top of guide, and the ~90% services gross margin sharpens the terminal profitability case. Against that, the cadence at scale is unproven, capex is running hot, and near-term service keeps sliding. At ~$93, down ~22% over the prior 30 days and off the $122 high, the stock discounts continued execution, but the de-rated entry plus a demonstrated catalyst is the most favorable risk/reward of our coverage. We would grow more constructive on a proven multi-satellite stacked launch and a repeatable second batch; we would reassess on a launch slip that pushes the 45-by-year-end target materially, a return of opex overruns, or undisciplined deployment of the "beyond the first 100" capital.

Thesis Scorecard Post-Earnings

We grade the standing thesis, pillar by pillar, against what this quarter's print and business update revealed. The upgrade trigger set at initiation and reaffirmed at Q3 (a Block 2 on orbit or a signed, quantified material government award) fired on both legs.

Overall: the thesis strengthened materially. Three bull pillars confirmed, the government pillar upgraded on a signed award, and the gating bear point (launch cadence) partially resolved for the first time in our coverage as BlueBird 6 retired the array-deployment risk. The residual risks are the cadence at scale and a fresh capex-discipline watch item, neither of which offsets the significance of a proven on-orbit deployment plus a contracted government prime.

Action: Upgrade to Outperform from Hold. We set an explicit, pre-committed upgrade trigger at initiation and again at Q3, and this quarter both legs fired: a Block 2 proven on orbit and a signed, quantified material government award. With the central technical risk demonstrated, the government leg contracted, revenue at the top of guide, the balance sheet funded past the full constellation, and a de-rated entry ~22% off the recent high, the risk/reward has tipped favorable. This is a measured upgrade, not a high-conviction one: the every-one-to-two-months cadence is still unproven at scale, and a material launch slip or a return of spending indiscipline would send us back to the sidelines. We would grow more constructive on a repeatable multi-satellite stacked launch; we own the story now because the constellation is finally being built on orbit, not just committed to on paper.