Firefly Aerospace Inc. (NASDAQ: FLY) — The Only Company That Landed Upright, Priced for a Medium-Lift Rocket That Hasn’t Flown

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice. The detailed analysis that follows it takes no position and carries no price target.

Verdict: HOLD / AVOID-chasing at ~$29. Accumulate-on-weakness only into the low-$20s / high-teens; do not pay up above ~$35. Not a short. Conviction: medium. Framing: a SpaceX-halo momentum vehicle wrapped around a genuine — but narrow — asset core, priced for the flawless execution of two things it has not yet done (fly a medium-lift rocket and turn a lunar cadence into a franchise).

Firefly is a better company than the space-SPAC wreckage around it and a worse stock than its +163% revenue line and Blue Ghost halo suggest. The core is real: it is the only commercial entity that has soft-landed on the Moon and stayed upright (its closest lunar peer, Intuitive Machines, tipped over on both attempts), and the October 2025 ~$547M SciTec acquisition bolted on a genuine national-security software business that is prime on an operational missile-warning program of record (FORGE) and holds a seat on the Golden Dome interceptor architecture. That combination — a proven lunar lander, a defense-AI prime, a responsive-launch niche the Space Force actually pays for, and a Northrop-Grumman-funded medium-lift program — is why this is not Astra and not a zero. But the price already capitalizes the option value of Eclipse (a 16-tonne rocket that has never left the ground), a monthly-lunar-cadence NASA program that exists mostly in Administrator slideware, and a “$3.2B Golden Dome” ceiling that is shared across twelve companies. Strip the optionality and the in-hand business — Alpha (a chronically unreliable small rocket that is only 16% of revenue), Blue Ghost, and SciTec — supports a fraction of the ~$4.3B enterprise value. At ~27x trailing / ~10x the $420–450M forward guide, you are paying a premium to the LUNR/Redwire “space-supplier” band (~4–9x forward) for a growth rate they lack, while funding ~$300M/yr of cash burn through serial dilution — and controlling sponsor AE Industrial Partners was the sole seller of 8.0 million secondary shares at $48 in late May (with a further 1.8M option), ~40% above today’s price, into the SpaceX-IPO scarcity bid, while no officer or director sold a share alongside it or has bought one in the open market.

What would move me: I flip constructive on a successful Eclipse maiden flight plus a firm, sized Golden Dome/CLPS-2.0 award (that converts the TAM slideware into contracted backlog) — at a better entry. I flip bearish on another Alpha or Blue Ghost mission loss, an Eclipse schedule slip into 2027+, or a third equity raise inside twelve months. The tag: the upright lander with a story bigger than its launch manifest. Own the execution, not the narrative — and not at 10x a number the company is still guiding you to believe.

📈 Stock Price Action — Five-Year Event Map

Firefly has no five-year history — it has an eleven-month one, and it has already round-tripped. The stock is an IPO-to-date creature: priced at $45 on August 6, 2025, it opened at $70, printed an intraday high of $73.80 on day one, and has spent every day since below the open. It cratered to $16.00 (November 21, 2025) — a ~78% drawdown from the peak — then clawed back to the low-$40s by April and spiked to $58.67 on May 27, 2026 on the SpaceX-IPO scarcity bid before fading to ~$28.90 (July 2, 2026). Current price sits ~61% below the day-one high, roughly at the IPO-day close, in a 52-week range of $16.00–$73.80 with a beta of ~3.2 — this is a high-amplitude thematic vehicle, not a compounder finding a level.

Cycle narrative. (1) The IPO was a hot-deal blowout — a 56% first-day pop over the $45 offer — that left the float underwater within 48 hours. (2) The subsequent slide to $16 tracked the April-2025 Alpha Flight 6 loss (FAA mishap investigation, no launch revenue) colliding with a de-rating of the entire 2025 space-IPO cohort. (3) The recovery was catalyzed by closing the transformative SciTec deal and the market re-appraising Firefly as a defense-software story, not just a rocket company. (4) The Q4/FY25 report and Blue Ghost’s landing anniversary drove a high-volume churn around $22–26. (5) Alpha’s clean Flight 7 return-to-flight (March) plus Space Symposium contract chatter and the building SpaceX-IPO scarcity trade squeezed the stock to $58. (6) Management and selling holders monetized that spike with a $48 follow-on — 8 million of the 12 million shares were secondary (insiders/AE Industrial), and the stock gapped down. (7) When the SpaceX debut passed, the whole listed-space basket deflated (LUNR, Redwire, Rocket Lab all faded together), carrying FLY back to the high-$20s even as it booked a real $144M NASA lunar-lander award on June 30. Price moves are Fact; attributed drivers are Interpretation.

1. Executive Summary

Firefly Aerospace is a Cedar Park, Texas space-and-defense company (founded 2017 as successor to the bankrupt Firefly Space Systems; ~1,400 employees; IPO August 2025) that has, in eleven months as a public company, transformed from a struggling small-rocket builder into a diversified, government-anchored space-and-defense-software platform — and been priced accordingly. It operates a portfolio spanning Alpha (a small, ~1-tonne-to-LEO responsive launch vehicle), Eclipse (a reusable 16-tonne medium-lift rocket in development with Northrop Grumman, not yet flown), Blue Ghost (a lunar lander — Mission 1 was the first fully successful commercial Moon landing, March 2025), Elytra (an in-space orbital-transfer/servicing vehicle), and an announced Ocula lunar-imaging service — plus, since the ~$547M October 2025 SciTec acquisition, a national-security software business (missile warning/tracking, AI data fusion) that is prime contractor on the operational FORGE program and holds a position on the Golden Dome interceptor architecture.

The financial profile is a classic hyper-growth, deep-loss story. FY2025 revenue was $159.9M, up 163% from $60.8M; Q1 2026 set a record at $80.9M, and management guides FY2026 to $420–450M (~+170%). Crucially, essentially all of the growth is Spacecraft Solutions — that segment went from $20.9M (FY23) to $131.2M (FY25) — while Launch revenue actually declined over the same span ($33.0M FY23 → $28.6M FY25). The revenue is now 84% “Spacecraft Solutions” (Blue Ghost + Elytra + SciTec software) and only 16% launch — Firefly is more a defense-and-lunar-services contractor than a launch pure-play, and its single largest customer (almost certainly NASA) was 59% of FY2025 revenue. But gross margin is only ~19–22% (a mix of fixed-price milestone and lower-margin cost-plus work), operating losses are widening (−$260.7M FY25), and the company burns ~$300M/yr of free cash flow. It is funded — $551.6M cash + a $305M undrawn revolver = ~$812M liquidity — but only because it raised ~$0.9B at IPO and another ~$192M in a May-2026 follow-on, and the ~1M-share/quarter dilution plus ~$50M/yr of stock comp are permanent features.

The bull case rests on genuine, differentiated assets — the only upright commercial Moon landing, a defense-AI prime with a program of record, a responsive-launch capability the Space Force uniquely values (VICTUS NOX, VICTUS DM), and a Northrop-funded path into the capacity-short medium-lift market — riding an enormous demand narrative (NASA’s Moon Base plan, Golden Dome, proliferated-LEO constellations). The bear case is equally concrete: extreme customer concentration (>86% of revenue from the top handful of accounts, ~100% government-adjacent), a chronically unreliable Alpha rocket, an unproven Eclipse carrying most of the equity’s optionality, ~$4.3B of enterprise value against $160M of trailing sales, a “controlled company” governed by a private-equity sponsor that just sold 8 million shares near the high, and a valuation (~10x forward sales) that leaves no margin for the launch and lunar-cadence slips that define this industry. This memo takes no position and sets no price target; it argues that Firefly is a real business whose equity is priced for outcomes it has not yet demonstrated.

2. Business Overview

Firefly Aerospace describes itself as an end-to-end “space and defense technology company” spanning launch, in-space operations, lunar delivery, and — post-SciTec — national-security software. The reality is a portfolio at very different maturities, only some of which generate meaningful revenue today. Understanding the revenue mix is the single most important orientation for this name, because the market narrative (“rocket company”) and the income statement (“defense-and-lunar-services contractor”) point in different directions.

Reporting structure and revenue mix. Firefly reports two revenue categories: Launch and Spacecraft Solutions. In Q1 2026, of $80.9M total revenue, Spacecraft Solutions was $67.6M (84%) and Launch was $13.3M (16%). Spacecraft Solutions is itself a bundle: the Blue Ghost lunar landers, the Elytra orbital vehicles, and — the fastest-growing piece — the SciTec national-security software business (FORGE, Golden Dome ground processing, SDA data-fusion). Launch is essentially Alpha, whose revenue is recognized at a point in time when a launch occurs — which, given Alpha’s low cadence and reliability record, makes it lumpy and small. Spacecraft Solutions revenue is recognized over time against contract milestones (percentage-of-completion), which management (correctly) frames as more predictable and recurring-like. The practical implication: the durable, growing revenue engine here is government software and lunar/spacecraft milestones, not rockets. (Fact: Q1’26 call, 2026-05-04; FY2025 10-K.)

The five product lines.

• Alpha — a two-stage, carbon-composite small launch vehicle carrying ~1,000 kg to LEO (and ~2,000 kg to suborbital/hypersonic-test trajectories). Its differentiator is responsiveness: Firefly executed the VICTUS NOX mission for the Space Force on a 24-hour launch call, and repeated the exercise with VICTUS DM. Alpha is operational but has a poor reliability record (see the Competitive Position section).
• Eclipse (MLV) — a reusable, scaled-up medium-lift vehicle developed in partnership with Northrop Grumman (which made a $50M equity investment and contributes Antares-derived technology), designed to deliver 16,000 kg to LEO (~16× Alpha’s mass to orbit) with reach to MEO/GEO/HEO/TLI. Powered by seven Miranda engines. It is in final development — flight-article tanks and interstage in test — but has not flown; and its path to market partly depends on Northrop’s internal/other bid approvals. Most of the equity’s launch-side optionality lives here.
• Blue Ghost — a lunar lander. Blue Ghost Mission 1 (March 2, 2025) was the first fully successful soft landing on the Moon by a commercial company — a $102.1M CLPS mission carrying 10 NASA payloads, completing all 17 objectives, operating 14 days of surface operations (and ~5 hours into lunar night). One caveat that undercuts the “launch company” narrative: BGM1 flew on a SpaceX Falcon 9, not on Firefly’s own Alpha — Firefly’s rocket cannot yet loft its own flagship spacecraft. Mission 2 (“Riders to the Dark,” a far-side landing, won for $130M) is assembling for a 2026 launch; Missions 3 and 4 are in design. This is the crown-jewel asset.
• Elytra — an orbital transfer/maneuver/servicing vehicle (with high-thrust Spectra engines) that first serves as a transfer stage and comms relay for Blue Ghost and then addresses space-domain-awareness, deorbit, and long-haul-communications missions (e.g., the DIU responsive-space work).
• Ocula — a lunar-imaging/remote-sensing service (high-resolution telescopes on Elytra, with an NVIDIA Jetson module for on-orbit AI processing) sold to government and commercial buyers.

How it makes money. Blended: (1) fixed-price, milestone-billed spacecraft and lunar-delivery contracts (Blue Ghost CLPS task orders, Elytra missions) recognized over time; (2) cost-plus and fixed-price defense-software contracts via SciTec (FORGE, Golden Dome, SDA) recognized over time; (3) point-in-time Alpha launch services; and (4) development-milestone revenue on Eclipse under the Northrop partnership. Backlog was ~$1.3–1.4B (Dec 2025 / Q1 2026), which management touts as cash-flow visibility — but the audited remaining performance obligations (the enforceable, contracted figure) were only $684.9M, less than half the promotional “$1.4B total backlog,” and ~81% of it sits with the top-five customers. Verdict: a coherent, vertically integrated space-and-defense portfolio, but one whose revenue today is dominated by government software and lunar milestones, with the launch business — the source of the “next SpaceX” narrative — still a small, unreliable minority of sales.

3. Industry Dynamics

Firefly straddles three distinct end-markets, each with different structural economics: launch, lunar/deep-space services, and national-security space software. Conflating them (as the “space stock” label does) obscures that they are not equally attractive.

Launch. The launch market is structurally bifurcated and brutal at the bottom. Small launch (Alpha’s segment) is a graveyard: dozens of entrants, chronic overcapacity, and a demand pool that SpaceX’s Transporter rideshare program has capped by offering dedicated-orbit delivery at a fraction of a dedicated small rocket’s price. Astra failed; Virgin Orbit failed; Rocket Lab’s Electron survives as the clear #2 only through scale, reliability, and a vertically integrated space-systems business layered on top. Small launch is a feature (responsiveness for national security) more than a profit pool. Medium/heavy launch (Eclipse’s target) is where the money and the shortage are: SpaceX’s Falcon 9 dominates, ULA’s Vulcan and Blue Origin’s New Glenn are ramping, and there is genuine, government-articulated demand for more Western providers to avoid single-source dependence — the opportunity Eclipse and Rocket Lab’s Neutron both chase. But medium-lift is capital-devouring and schedule-cursed: Neutron has slipped repeatedly, and no new US medium-lift entrant has yet reached orbit. The capital cycle here (Marathon lens) is at peak investment with returns still hypothetical.

Lunar & deep-space services. This is a government-created market — essentially NASA’s Commercial Lunar Payload Services (CLPS) program and its successors, plus emerging cislunar national-security demand. It is real and growing: CLPS 1.0’s budget rose from $2.6B to $4.2B, a CLPS 2.0 (~$6B) is anticipated, and NASA’s “Moon Base” architecture implies (per Firefly’s framing) a ~$20B program over seven years with monthly robotic landings. The structural catch: these are fixed-price government service contracts with historically punishing margins — the CLPS model has produced negative-to-low-single-digit gross margins across the field, and every lunar contractor has taken losses on missions. Demand visibility is high; profitability is not. And demand is entirely dependent on the NASA budget and Artemis political will.

National-security space software. The SciTec business sits in the most attractive of the three markets: sticky, program-of-record defense software (missile warning/tracking, data fusion, command-and-control) with 45 years of proprietary algorithms, incumbency on FORGE, and exposure to the enormous Golden Dome missile-defense build-out. This is higher-margin, recurring, and moated by clearances, accreditation, and switching costs — genuinely the best profit pool Firefly touches, and the reason the SciTec deal re-rated the stock.

Regulatory & demand structure. Launch requires FAA licensing (and a mishap investigation halts flights, as Alpha’s April-2025 anomaly did); everything is ITAR-controlled; and the customer base is ~100% US-government-adjacent (NASA, USSF, SDA, NRO, DoD, and prime contractors Lockheed/Northrop/L3Harris). That yields demand durability but also budget/appropriations risk and procurement lumpiness. Verdict: a mixed industry verdict — the defense-software and (subsidized) lunar markets are structurally attractive and growing; small launch is structurally poor; medium launch is potentially lucrative but unproven and capital-cursed. Firefly’s fortunes hinge on the two good markets (defense software, lunar) subsidizing its push into the hard one (medium launch).

4. Competitive Position

The moat question must be answered product-by-product, because Firefly’s competitive standing ranges from genuine leadership (lunar landing) to subscale also-ran (small launch).

Lunar — a genuine, if narrow, edge. Blue Ghost Mission 1 is the strongest competitive asset Firefly owns: it is the only commercial lunar lander to soft-land fully upright and complete a full surface mission. Its nearest peer, Intuitive Machines (LUNR), landed but tipped over on both IM-1 and IM-2; Astrobotic’s Peregrine never reached the Moon. In a CLPS field where NASA is buying reliability for increasingly expensive payloads, a demonstrated upright landing with 14 days of operations is a real, evidence-backed differentiator — and it showed up in the June-30 $144M CLPS award. The moat mechanism is demonstrated technical capability + vertical integration (Firefly builds its own avionics, composites, and engines), which lowers cost and de-risks the next lander. It is durable for now, but not permanent: SpaceX (Starship HLS), Blue Origin (Blue Moon), and a re-tooling Intuitive Machines all threaten, and higher payload-mass classes favor larger players.

Defense software (SciTec) — the best moat, bought not built. SciTec’s incumbency as prime on FORGE (an operational missile-warning program of record, battle-tested in the Iran conflict) plus 45 years of accredited algorithms creates classic government-software stickiness: switching costs, security clearances, and integration lock-in. Its Golden Dome seat (one of twelve awardees on contracts totaling up to $3.2B collectively) is optionality, not yet a sized moat. This is the highest-quality competitive position in the portfolio — but Firefly acquired it for ~$547M rather than building it, so the durability question is really “can Firefly retain SciTec’s talent and win recompetes,” not “does Firefly have a technology edge.”

Launch — subscale challenger, thin moat. Alpha’s record is 5 successful launches out of 7 — and 2025 alone brought two anomalies (the April Flight 6 loss and a September 29 test-stand event at Briggs that damaged the stand). “First U.S. company to reach orbit on its second attempt” (Flight 2, 2022) is a real credential, but a 5-of-7 record is a thin reliability moat for a business selling reliability. In small launch, Alpha competes against Rocket Lab’s Electron (the reliable, high-cadence #2) and, brutally, against SpaceX Transporter rideshare on price. Alpha’s only real differentiator is responsive/tactical launch — the 24-hour-call VICTUS NOX/DM capability the Space Force pays a premium for — a genuine but small, government-specific niche. In medium launch, Eclipse will face Falcon 9, Vulcan, New Glenn, and Neutron; it has the Northrop partnership and a reusable design, but it has not flown, and “hasn’t reached orbit” is the entire history of failed medium-lift hopefuls. The moat here is prospective, not demonstrated.

Vertical integration is the cross-cutting structural advantage Firefly claims — common carbon-composite structures, all-in-house engines built on a patented tap-off cycle (five granted US patents; Alpha’s Reaver/Lightning, Eclipse’s Miranda/Vira, Elytra’s Spectra), avionics, and co-located rocket-and-spacecraft factories — which, if executed, lowers unit cost and shares R&D across products. It is a credible cost-side thesis, but the −18.7% gross margin Alpha produced in 2024 shows integration has not yet translated into economics. Verdict: Firefly has one demonstrated moat (upright lunar landing), one acquired moat (defense software), and one aspirational moat (low-cost vertically integrated launch). That is materially better than the no-moat space-SPAC cohort — but it is not the durable, self-reinforcing advantage the ~$4.3B valuation implies, and the crown-jewel lunar edge is the narrowest and most contestable of the three.

5. Growth History and Forward Opportunities

History. Revenue grew from $55.2M (FY2023) to $60.8M (FY2024, +9%) to $159.9M (FY2025, +163%), then accelerated to a record $80.9M in Q1 2026 (+45% YoY). The step-change is real but its composition matters: Spacecraft Solutions grew from $20.9M (FY23) to $131.2M (FY25) while Launch revenue fell ($33.0M → $28.6M) — so the surge is driven entirely by SciTec software (FORGE ramp, Golden Dome), Blue Ghost/spacecraft milestones, and acquired revenue, not by a scaling launch cadence (Q1’26 launch was just $13.3M on the single Alpha Flight 7). On the demand side, Firefly does hold multi-launch framework agreements — Lockheed Martin (up to 25 missions over five years) and L3Harris (up to 20 Alpha launches, 2–4/yr from 2027–2031) — but these are option-heavy manifests, not firm near-term revenue.

Forward drivers (management’s framing, treated as hypothesis).

1. FY2026 guide $420–450M (~+170%) — the near-term proof point; a full year of SciTec, the FORGE/Golden Dome ramp, Blue Ghost Mission 2 milestones, and 3–4 Alpha launches.
2. Lunar cadence — NASA’s “Moon Base” acceleration (monthly robotic landings; CLPS 1.0 → 2.0 → larger 3-tonne and 8-tonne landers post-2030). Firefly is scaling a 4x-larger clean room to templatize Blue Ghost production. If the NASA budget materializes, this is the largest single opportunity — but it is the most policy-dependent.
3. Golden Dome / defense software — SciTec’s transferable AI algorithms position it for space-based-interceptor ground processing; a $109M FORGE ECP was booked in Q1’26. Sizable but not yet contracted at scale.
4. Eclipse entry into medium-lift — the capacity-short, high-value market; maiden flight is the swing factor for the launch thesis.
5. International/responsive launch — a Swedish launch pad (Esrange), a Seagate sea-launch platform concept, allied-nation demand for sovereign launch.
6. Elytra/Ocula cislunar services — SDA, DIU, and NVIDIA-enabled on-orbit processing.

Quality of growth. Mixed. The demand signals are genuine and unusually strong (this is not a company searching for a market). But the growth is (a) heavily government-funded and appropriations-dependent, (b) substantially acquired (SciTec) rather than organic, © low-margin on the lunar/cost-plus side, and (d) reliant on execution milestones (Eclipse first flight, Alpha cadence, Blue Ghost missions) that this industry routinely misses. Verdict: high-magnitude but medium-quality growth — real and demand-backed, but government-dependent, margin-thin, partly bought, and execution-contingent. The $420–450M guide is the credibility test; hitting it validates the trajectory, missing it (via an Alpha or Eclipse slip) would puncture the multiple.

6. Financial Quality

Firefly’s financials are those of a well-funded, fast-scaling, deeply unprofitable hardware company — with several quality-of-earnings wrinkles worth isolating.

Revenue and margins. FY25 revenue $159.9M (+163%); gross profit $30.7M at a 19.2% gross margin — a swing from FY24’s negative −18.7% (Alpha failures and inventory/write-off drag) but well below FY23’s 48.2% (a smaller, more favorable mix). Q1’26 GAAP gross margin was 21.6%, down from 27.7% in Q4’25 on “a higher mix of cost-plus program contracts.” The margin signal is that as SciTec cost-plus and CLPS lunar work scale, blended gross margin sits in the low-20s%, not the 40s% — this is a government-contractor margin profile, not a software or a mature-launch one.

Operating losses and burn. FY25 operating loss was −$260.7M (R&D $200.1M — a staggering 125% of revenue, funding Eclipse/Alpha Block 2/lunar — plus SG&A $91.2M). Net loss −$298.3M; EPS −$4.83. Operating cash flow was −$204.9M, and with $32.8M capex, free cash flow was −$237.8M. Q1’26 FCF was −$78.9M (~$315M annualized), with adjusted EBITDA of −$64.7M. This is a ~$300M/yr cash-burn machine whose burn is rising as it invests into the ramp.

Balance sheet & liquidity. Post-IPO and follow-on, the balance sheet is genuinely solid for now: at Q1’26, $551.6M cash + short-term investments, ~$53M total debt (the $260M revolver was drawn and repaid; it is now upsized to $305M and undrawn), net cash ~$299M, and total liquidity of $811.6M. Deferred revenue/customer advances of ~$146M fund working capital (a negative cash-conversion cycle — customers pay ahead of milestones, a genuine positive). Runway at the current burn is roughly 2–2.5 years before another raise — adequate, not comfortable, for a company still pre-Eclipse.

Quality-of-earnings flags. (1) Goodwill $453M + intangibles $160M (~$886M combined, ~40% of the $1.49B balance sheet) sit from the SciTec deal — an impairment watch-item on a unit owned only two months at year-end. (2) The FY25 net loss line contains large non-cash noise in both directions: “total other expense, net” was a ~$74.8M net expense (a −$50.3M warrant-fair-value remeasurement and a −$30.4M loss on debt extinguishment, partly offset by net interest and an $8.4M contingent-liability gain), while the loss was reduced by a one-time, non-cash $37.1M income-tax benefit from establishing SciTec’s deferred-tax liability. The cleaner read is the operating loss (−$260.7M) and the cash burn, not GAAP net loss. (3) Stock-based comp is escalating fast — $17.8M for FY25 but $12.5M in Q1’26 alone (~$50M annualized run-rate post-IPO), a real, recurring dilution cost the non-GAAP figures add back. (4) The pre-IPO structure carried redeemable convertible preferred (carrying value $759.6M, liquidation preference $1,227.2M) that converted into ~105.8M common shares at IPO (driving −$918M equity pre-IPO to positive after); the accumulated deficit is $1.12B. (5) Revenue is milestone/percentage-of-completion with ~90% of contract value typically collected before delivery — and deferred revenue actually drew down $9.9M in Q1’26, meaning reported revenue was partly funded by prior customer prepayments. Share count grows ~1M/quarter. Verdict: economics are not yet improving with scale in any way that matters — gross margin is stuck in the low-20s%, the operating loss is widening, and the business is entirely dependent on external capital. The balance sheet buys time; it does not answer the profitability question, which remains years and an Eclipse-cadence away.

7. Capital Allocation

Capital allocation at Firefly is, appropriately for its stage, about deploying raised capital into growth — but the pattern already reveals management’s priorities and a few alignment flags.

Sources. The August 2025 IPO (22.2M shares at $45, including a fully exercised 2.9M over-allotment; a 1-for-3.2544 reverse split was effected at pricing) grossed ~$998.6M and netted ~$932.5M. Proceeds first repaid a punitive prior term loan — $136.1M at a 13.875% rate (Term B stepping to 19.135% in July 2026), insurance-backed — incurring an $11.4M prepayment premium and a $30.4M loss on extinguishment (AE Industrial itself held ~$21.1M of that loan and was repaid). The May-2026 follow-on added ~$190M of primary proceeds (4.0M company shares at $48). A revolver ($125M → $260M → $305M) at SOFR + 3.00% backstops liquidity (covenants: min liquidity $120M, min FCF −$325M).

Uses. (1) R&D: $200M in FY25 (125% of revenue) — the dominant use, funding Eclipse development, Alpha Block 2, and lunar production. For a pre-scale hardware company this is defensible, but it is the reason for the burn. (2) The ~$547M SciTec acquisition (closed October 31, 2025) — $277.4M cash (funded by a $260M revolver draw, since repaid) plus $269.6M in stock — the defining capital-allocation decision. Strategically it is the best thing management has done: it diversified Firefly out of pure launch into higher-quality, higher-margin, sticky defense software and materially re-rated the equity. It created ~$453M goodwill + ~$160M intangibles (~$613M, an impairment-exposed base for a unit owned two months at year-end), but the strategic logic is sound. (3) Capex ~$33M FY25 (clean rooms, test stands, factory expansion) — scaling for lunar cadence. Note that the $35.6M of FY25 “preferred dividends” that depressed EPS-to-common was non-cash accretion, not a cash outflow (the only cash preferred dividend was a $5.0M IPO-closing payment). No common dividend or buyback (the revolver prohibits them).

Alignment & governance flags. First, Firefly is a “controlled company”: AE Industrial Partners controls a majority of the voting power (Director Nomination Agreement + voting proxies), sits on a classified (staggered) board where 5 of 8 directors are AE employees, may designate up to 55% of the board, and there is no say-on-pay (emerging-growth-company exemption) — minority holders have limited say. An AE consulting agreement pays ~$2.4M/yr. Second, AE Industrial was the sole selling stockholder in the May-2026 follow-on — 8.0M shares at $48 plus a 1.8M option (~$384–470M), cutting its stake from ~36.7% to ~31.7% near the ~$58 local peak, ~40% above today’s price. No officer or director sold alongside it, and there has been no open-market insider buying (zero code-P) to offset — the capital-markets tell is sponsor distribution, not conviction. Third, CEO Jason Kim’s FY2025 total compensation was $38.2M — almost entirely a $36.05M IPO RSU grant (CFO Darren Ma and COO Ramon Sanchez ~$11.3M each) — a large number for a pre-profit company, though front-loaded IPO grants are common; Kim owns just ~1.8% of the stock. Verdict: the strategic capital allocation (SciTec, R&D into differentiated assets) is rational and arguably value-creating; the governance and alignment picture (sponsor control, a classified AE-dominated board, no say-on-pay, secondary selling into the peak, no insider buying, outsized comp) warrants real skepticism. Management is building genuine assets — and the controlling owner is already taking chips off the table.

8. Changes and Headwinds — Last Two Years

The last ~24 months contain nearly the entire public history of this company and several thesis-defining events:

• Blue Ghost Mission 1 lands on the Moon (March 2025) — the first fully successful commercial lunar landing; the single most important credibility event in Firefly’s history and the foundation of its lunar franchise.
• Alpha Flight 6 anomaly (April 29, 2025, Vandenberg) — a launch failure that lost the mission, triggered an FAA mishap investigation, and grounded Alpha until August 26, 2025; a stark reminder of launch-reliability risk.
• IPO (August 6, 2025) at $45 (opened $70) — raised ~$0.9B and made Firefly a public, high-visibility space name.
• SciTec acquisition (~$547M, closed October 31, 2025) — the transformative deal that added a defense-software prime and re-rated the equity from “rocket company” to “space-and-defense platform.”
• Alpha Flight 7 successful return-to-flight (March 2026) — deployed a Lockheed demonstrator, validated Block 2 subsystems; restored launch credibility.
• VICTUS DM responsive-launch demonstration (Q1 2026) — second tactically responsive mission, reinforcing the national-security niche.
• Golden Dome seat + $109M FORGE ECP (Q1 2026) — SciTec positioned on the space-based-interceptor architecture (one of twelve awardees) and expanded its operational missile-warning contract.
• Follow-on offering (May 28, 2026, $48) — 4M primary + 8M secondary shares; insiders/AE Industrial distributed into the SpaceX-IPO scarcity bid.
• $144M NASA CLPS lunar-lander award (June 30, 2026) — a concrete lunar-cadence win alongside Astrobotic and Intuitive Machines.
• NVIDIA partnership for Ocula on-orbit AI processing; COO Ramon Sanchez hired (Q4’25) to industrialize production.

Headwinds: the SpaceX-IPO-driven space-basket unwind (June 2026 “space gloom”) pressured the whole cohort; Alpha’s reliability remains an overhang; Eclipse’s maiden flight keeps slipping toward late-2026/2027; and the entire demand narrative is hostage to the NASA and defense appropriations cycle. Verdict: the two-year record strengthens the business thesis (a landed Moon mission, a transformative acquisition, a return-to-flight, real contract wins) while weakening the near-term stock thesis (a failure reminder, serial dilution, insider distribution, and a valuation that ran ahead of the fundamentals before fading).

9. Risk Analysis (Risk Matrix)

The dominant risks are execution (launch/Eclipse), concentration/budget dependence, and valuation — a combination in which any single miss (an Alpha failure, an Eclipse slip, a NASA budget cut, or simply a continued space-basket de-rating) can compress the multiple faster than the fundamentals deteriorate, given how much of the price is optionality.

10. Valuation Discussion (Embedded Expectations)

Firefly cannot be valued on earnings (deeply negative), cash flow (−$300M/yr), or book value (dominated by acquisition intangibles). The only workable frame is EV/sales against space peers plus a reverse-expectations check — and, because this is a fresh IPO, there is no own-history percentile (AZI/FactorsToday return nothing) to anchor “cheap vs. its own range.”

Where the multiple sits. At ~$28.90 on ~160M shares, market cap is ~$4.6B; net cash ~$0.3B gives EV ~$4.3B. Against FY25 sales of $159.9M that is ~27x trailing; against the $420–450M FY26 guide it is ~10x forward. The peer set:

*LUNR forward multiple is low only because its FY26 revenue jump is largely acquired (Lanteris), not earned. (Peer figures per each company’s mid-2026 public filings.)

Read. On trailing sales FLY (~27x) is far cheaper than RKLB (~90x) and vastly cheaper than pre-revenue ASTS, but ~2–3x the trailing multiple of its closest scale/character peers LUNR (~12x) and RDW (~9x). On forward sales the picture inverts favorably: FLY’s ~10x forward reflects the fastest growth in the cohort (+163% vs. RKLB ~37%, RDW +10%, LUNR negative organically), so the multiple compresses fastest if the guide is hit organically (and, unlike LUNR, most of FLY’s forward growth is not purely acquired). The honest characterization: FLY is not cheap, but it is more defensibly priced than RKLB and dearer than the “space-supplier” band — a valuation that is a bet on a growth rate the cheaper peers do not have.

Embedded expectations (reverse check). To justify ~$4.3B EV at a ~12% cost of capital, the enterprise must compound to roughly $7.5–8B by 2030 merely to earn its cost of capital, more to deliver a growth-equity return. At a generous 8x exit EV/sales (RKLB-lite), that back-solves to ~$950M–$1.0B of 2030 revenue to break even on capital and ~$1.3B+ to earn a real return — a ~6–8x from today’s $160M, i.e., a sustained ~40–45% revenue CAGR plus a margin turn toward self-funding. That is demanding but — unlike RKLB, where the reverse-DCF barely works even in a bull case — not obviously impossible given the FY26 guide already implies ~$435M and the lunar/defense demand backdrop.

Scenarios (ranges, no target).

• Bear: Alpha stays subscale, Eclipse slips past 2027, NASA budget disappoints → ~$400M revenue by 2030 at 6–8x → ~$2.4–3.2B EV (below today); the multiple de-rates toward the LUNR/RDW band and dilution impairs the equity.
• Base: Eclipse flies, CLPS cadence continues, SciTec/defense scales → ~$1B revenue by 2030 at ~8x → ~$8B EV — roughly justifies today’s EV with a modest return, contingent on clean execution.
• Bull: Firefly becomes the credible #3 Western launch provider and monetizes a lunar-services franchise (Blue Ghost’s upright-landing edge) → ~$2B+ revenue by 2030 at 8–10x → ~$16–20B EV.

What the market is underwriting. At ~10x forward, the market is pricing the base case with execution optionality — i.e., that Eclipse flies, the lunar cadence is real, and SciTec/Golden Dome scales. It is not pricing a catastrophic Alpha/Eclipse failure, and it is not fully crediting the bull. The valuation is a momentum/story repricing riding the SpaceX halo, with a fundamental floor (Alpha + Blue Ghost + SciTec + backlog) that supports materially less than $4.3B; the balance is Neutron-style optionality applied to unproven medium-lift and lunar cadence.

11. Variant Perception

Consensus view. Post-SpaceX-IPO, the sell-side and momentum crowd frame Firefly as a “next-generation, vertically integrated space-and-defense platform” — the rare space name with a landed Moon mission, a defense-software prime, and a Northrop-backed medium-lift rocket — riding secular tailwinds (Artemis/Moon Base, Golden Dome, proliferated-LEO). KeyBanc’s Overweight and $50 price target (June 2026) typify the bull framing: a multi-market compounder early in a multi-year demand super-cycle.

Strongest bull case. Firefly is the highest-quality asset in the listed-space cohort below Rocket Lab: the only upright commercial Moon landing (a real CLPS edge over a fumbling LUNR), a genuine defense-AI prime (FORGE) with Golden Dome optionality, a uniquely valued responsive-launch niche, and a Northrop-funded path into the capacity-short medium-lift market — growing +163% with $1.3B of backlog and ~$800M of liquidity. If Eclipse flies and the NASA/defense demand converts to contracts, the base/bull scenarios ($8–20B EV) leave meaningful upside, and today’s ~10x forward is cheap for the growth.

Strongest bear case. Firefly is a ~$300M/yr cash-incinerating, government-dependent contractor with >86% customer concentration, a chronically unreliable small rocket that is only 16% of revenue, low-20s% gross margins, and ~40% of its assets in acquisition goodwill — trading at ~27x trailing sales, ~10x a guided forward number, with most of its EV in an unflown rocket and slideware TAM. It is a “controlled company” whose PE sponsor just sold 8M shares near the peak with no insider buying to offset. High beta (~3.2) makes it a SpaceX-halo momentum vehicle that already fell 38% when the scarcity bid unwound; another Alpha/Eclipse miss or NASA budget cut compresses the multiple toward the LUNR/RDW band (implying a materially lower stock).

The 3–5 assumptions that matter most. (1) Eclipse reaches orbit on a credible schedule (bull) vs. slips/fails (bear). (2) The FY26 $420–450M guide is hit organically — the near-term credibility test. (3) NASA’s lunar-cadence and Golden Dome demand convert into sized, contracted backlog rather than remaining agency slideware. (4) Alpha and Blue Ghost avoid another mission loss. (5) Blended gross margin inflects toward the 30s% (self-funding path) vs. stays stuck in the low-20s%.

Factor-positioning read. Firefly is too new for the FactorsToday model (no 252-day history) and has no AZI own-history valuation percentile — but the tape tells the story directly: a ~3.2 beta, an eleven-month round-trip from $73.80 to $16 to $58 to $29, and a demonstrated −38% move when the SpaceX-IPO scarcity trade unwound mark this as a high-amplitude thematic-momentum vehicle, not a fundamentally anchored equity. The variant-perception edge is that consensus is trading the narrative (Moon, Golden Dome, “next SpaceX”) while the in-hand fundamentals (a subscale rocket, low-margin lunar contracts, an acquired software business) support a fraction of the price — so the stock is likely to remain hostage to sector sentiment and execution binaries rather than to compound smoothly. Where consensus is most offsides: treating the enormous addressable TAM (NASA Moon Base “$20B,” Golden Dome “$3.2B”) as if it were Firefly’s contracted revenue, when the company itself guides to $435M.

12. Fact vs. Interpretation Table

13. Open Questions

1. What is SciTec’s standalone revenue, margin, and organic growth? Management declines to break it out — yet it is the fastest-growing, highest-quality piece. Without it, the organic Firefly growth rate is unknowable.
2. What is Eclipse’s realistic maiden-flight date and total remaining development cost? “Late 2026/2027” and the burn trajectory both hinge on this.
3. What are Blue Ghost / CLPS gross margins? Management won’t disclose spacecraft segment margins; the industry precedent is negative-to-low. Are the lunar contracts profitable or loss-leaders for cadence?
4. How much of the $1.3B backlog is firm vs. optional, and what is its margin? ~81% sits with five customers; conversion timing and profitability are opaque.
5. What is Firefly’s actual (vs. ceiling) Golden Dome award? The “$3.2B” is shared across twelve firms; Firefly’s sized share is undisclosed.
6. When is the next equity raise, and at what dilution? ~2–2.5-year runway at current burn, but a growth acceleration (or an Eclipse overrun) shortens it.
7. Will AE Industrial continue distributing? The May secondary was the first; lockups and further sponsor selling are an overhang.
8. What is normalized gross margin at scale? The low-20s% blend must inflect for the base case to work — is that a mix question or a structural ceiling?

14. What Must Be True

For the bull case (base/bull scenarios, ~$8–20B EV):

1. Eclipse reaches orbit on a credible schedule and ramps into the capacity-short medium-lift market. Falsification: an Eclipse maiden-flight failure, or a schedule slip beyond 2027, collapses the launch-side optionality that underpins most of the EV.
2. The FY26 $420–450M guide is hit (largely organically) and 2027 accelerates, with blended gross margin inflecting toward the 30s%. Falsification: a material FY26 revenue miss or gross margin stuck in the low-20s% signals the growth is bought and the model doesn’t self-fund.
3. NASA lunar cadence and Golden Dome convert to sized, contracted backlog, and Alpha/Blue Ghost avoid further mission losses. Falsification: NASA budget cuts, a lunar-cadence delay, or another launch failure turns the TAM slideware into a broken narrative.

For the bear case (below today, ~$2.4–3.2B EV):

1. Alpha stays subscale and Eclipse slips/fails, leaving Firefly a low-margin government-services contractor rather than a launch platform. Falsification: a clean Eclipse debut plus rising Alpha cadence proves the launch thesis.
2. Cash burn forces repeated dilution while gross margin stays in the low-20s%, and AE Industrial keeps selling. Falsification: burn narrows toward breakeven on the revenue ramp and the sponsor stops distributing (or insiders buy).
3. The multiple de-rates toward the LUNR/RDW band (~4–9x forward) as the SpaceX-halo unwinds. Falsification: the stock holds a premium forward multiple through the sector’s volatility, validating a genuine quality differentiation.

The analysis above takes no position and contains no price target; valuation is discussed only as embedded expectations and scenarios. The single, deliberately fenced exception is the Claude's Take block at the top, which is the author’s own independent opinion and general information only, not investment advice.

APPENDIX A — Standard Diligence Questionnaire

Firefly Aerospace Inc. (NASDAQ: FLY) — as of 2026-07-04

Supplemental to the research memo. Answers grounded in the FY2025 10-K, Q1 2026 10-Q, DEF 14A, S-1, 8-Ks, the Q1 2026 earnings call, and peer reports. Fact / Interpretation / Assumption labeled where it matters.

General

What thoughtful questions have other investors asked? (1) How much of the growth is organic vs. acquired? — SciTec (bought Oct 2025) drove much of the FY25/FY26 step-up; management won’t break out its standalone revenue (Fact). (2) When does Eclipse fly, and what does it cost to finish? — the swing factor for the launch thesis (Open Question). (3) Are the CLPS/lunar contracts actually profitable? — the industry precedent is negative-to-low margin (Interpretation). (4) Why is launch revenue flat-to-declining if this is “the next SpaceX”? — Launch fell from $33.0M (FY23) to $28.6M (FY25) (Fact). (5) Is the “$1.4B backlog” real? — audited RPO is only $684.9M (Fact).

Cyclicality & Earnings Nature

Cyclical high or low? Neither in the classic sense — Firefly is a pre-profit, pre-scale growth company; “earnings” are deeply negative (−$298M FY25) and driven by internal investment (R&D 125% of revenue), not a cycle (Fact). The relevant “cycle” is the government appropriations cycle (NASA/Artemis, DoD/Golden Dome), which is currently at a high in political enthusiasm (Interpretation).

External environment or internal actions? Both: revenue depends on winning government contracts (external) but is gated by execution — launches, landings, milestones (internal). The April-2025 Alpha failure (internal execution) directly cost launch revenue and grounded the fleet (Fact).

How stable are revenues? Low stability — milestone/percentage-of-completion and point-in-time launch revenue is lumpy; ~35% of FY25 revenue landed in Q1’25 on the Blue Ghost Mission 1 delivery (Fact). Deferred revenue ($146M) provides some visibility but drew down $9.9M in Q1’26 (Fact).

Outlook / market size. Very large addressable markets (company cites a $1.8T space economy by 2035, a $32B launch market, a $7B spacecraft market) but these dwarf Firefly’s $160M revenue and $435M guide — the TAM is aspirational relative to realized revenue (Interpretation). Growing, domestic-and-allied, government-anchored.

Business Quality & Competitive Moat

Industry more or less competitive? Launch is intensely competitive and structurally poor at the small end (SpaceX Transporter caps pricing; Astra/Virgin Orbit failed); medium-lift is lucrative but crowded and unproven; lunar and defense-software are less crowded (Interpretation).

How profitable is the business? It is not — ROIC/ROE are meaningless (negative); gross margin is ~19–22%, operating margin −163% FY25 (Fact). Economics do not yet improve with scale.

How profitable is the industry / barriers to entry? Barriers are high (capital, engineering, clearances, ITAR, flight heritage) but so are the losses — every listed peer (RKLB, LUNR, RDW, ASTS) burns cash. Firefly’s genuine barriers: the only upright commercial Moon landing, a defense-software incumbency (FORGE), and responsive-launch clearances (Fact/Interpretation).

Easily understood? Moderately — five product lines at different maturities plus an acquired software unit make it more complex than a single-product company.

Undermined by foreign low-cost labor? No — ITAR/national-security work is inherently domestic; the risk is foreign competition (Chinese launch, European medium-lift), not labor arbitrage.

Do brands matter? In the sense of flight heritage and reliability reputation, yes — “first commercial Moon landing” is a marketable credential; consumer branding is irrelevant (Interpretation).

Nature of competition / switching costs. Government software (SciTec) has real switching costs (accreditation, integration); launch and lunar are re-competed contract-by-contract with lower stickiness (Interpretation).

Financial Condition & Balance Sheet

Assets not fully on the balance sheet? The flight heritage / IP / patented engine cycle (5 patents) is not capitalized; conversely, ~$886M of goodwill+intangibles (~40% of assets) is on the balance sheet from SciTec and is impairment-exposed (Fact).

Off-balance-sheet liabilities? Operating/finance leases (~$20M); a $305M undrawn revolver (available, not drawn); ~90%-collected-in-advance customer contracts create deferred-revenue obligations (Fact). No unusual off-balance-sheet structures identified.

How conservative is the accounting? Mixed — percentage-of-completion revenue relies on management cost-to-cost estimates (aggressive-capable); the SciTec purchase accounting created large intangibles; non-GAAP add-backs (SBC $50M/yr, one-time items) are substantial (Interpretation). Auditor (Grant Thornton) issued an unqualified opinion (Fact).

CapEx-hungry? Yes, structurally — space hardware requires factories, test stands, clean rooms; capex ~$33M FY25 and rising, on top of R&D of $200M (Fact).

Capital Allocation & Management

FCF generation / use / philosophy. Negative FCF (~−$300M/yr); “use of capital” is funding growth (R&D + the SciTec acquisition + capex) via raised equity/debt. Philosophy is invest-for-scale, standard for the stage (Fact).

Significant acquisitions? Yes — SciTec (~$547M, Oct 2025), the transformative, arguably value-creating deal; a small undisclosed Space-ng (per June-2026 news, not in filings) (Fact / Open Question).

Buying back shares? No — prohibited by revolver covenants; the company is a net issuer (~1M shares/quarter dilution + a May-2026 primary raise) (Fact).

Issuing shares to insiders? Large IPO equity grants (CEO $36M RSU grant); and the controlling sponsor AE Industrial sold 8–9.8M secondary shares at $48 (Fact).

Compensation policy. CEO FY25 comp $38.2M (mostly one-time IPO RSUs); no say-on-pay (EGC exemption); classified board dominated by AE Industrial (Fact). Alignment is weak on governance, though the CEO does hold ~1.8% (Interpretation).

Motivations of management. Build a diversified space-and-defense platform; the controlling PE sponsor (AE Industrial) is motivated to monetize its ~2022 investment (already selling) (Interpretation).

Valuation & Market Data

ADR / MLP / K-1? No — a Delaware C-corp, single class of common stock, Nasdaq-listed. Not an ADR/MLP; no K-1.

Dividend policy. None on common (and prohibited by covenants); the FY25 “$35.6M preferred dividend” was non-cash accretion on now-converted preferred (Fact).

How profitable? Unprofitable (see above).

Net income vs. cash from operations. Both deeply negative; net loss −$298M FY25 vs. OCF −$205M — the gap is non-cash (SBC, warrant remeasurement, D&A, tax benefit) (Fact).

Risks & Downside

What would cause the stock to decline? An Alpha or Blue Ghost mission failure; an Eclipse slip/failure; a NASA/DoD budget cut; a further equity raise; continued SpaceX-halo sector de-rating; or simply the ~10x-forward multiple compressing toward the LUNR/RDW band (Interpretation).

Risk of catastrophic loss? Moderate — a single high-profile mission failure (lunar or Eclipse maiden flight) could puncture the thesis and the multiple simultaneously; the ~3.2 beta amplifies moves (Interpretation).

Chance of total loss? Low near-term — ~$740M pro-forma cash, ~$1.3B backlog, real contracts, and a controlling sponsor make a zero unlikely absent a catastrophic reputational event; but the equity is a high-variance, dilution-exposed instrument (Interpretation).

Recent News & Events

Has the environment changed recently? Yes — the June-2026 SpaceX IPO re-rated then de-rated the entire listed-space complex; Firefly rode it to $58 and back to $29 (Fact). NASA’s “Moon Base” acceleration and the Golden Dome build-out are recent, favorable demand signals; a $144M NASA CLPS award landed June 30, 2026 (Fact).

Significant acquisitions? SciTec (Oct 2025) — see above.

Accounting-policy changes? None material beyond the SciTec purchase-accounting and the IPO-driven conversion of preferred/warrants (Fact).

Recent changes — markets, facilities, management? New COO (Ramon Sanchez, Dec 2025) to industrialize production; a 4x-larger clean room for lunar cadence; new/expanding launch sites (Wallops, Esrange Sweden, planned Cape Canaveral); an NVIDIA partnership for Ocula on-orbit AI (Fact).

APPENDIX B — Source Appendix

Firefly Aerospace Inc. (NASDAQ: FLY) — as of 2026-07-04

Primary sources first. All SEC filings accessed via EDGAR (CIK 0001860160) and mirrored locally to output/FLY/sources/. Quantitative data cross-checked against ROIC.ai and the AZI feeds; reconciled to filings where they differ (the filing is authoritative).

Primary — SEC filings (EDGAR, CIK 0001860160)

1. FY2025 Form 10-K (filed 2026-03-20; period ended 2025-12-31) — business, risk factors, MD&A, financial statements, revenue segmentation (Launch vs. Spacecraft Solutions), backlog/RPO, customer concentration, SciTec purchase accounting, debt, going-concern basis. https://www.sec.gov/Archives/edgar/data/1860160/000119312526116309/fly-20251231.htm
2. Q1 2026 Form 10-Q (filed 2026-05-04; period ended 2026-03-31) — Q1’26 results, balance sheet, liquidity, deferred revenue, warrant liability, share count. https://www.sec.gov/Archives/edgar/data/1860160/000186016026000009/fly-20260331.htm
3. DEF 14A Proxy (filed 2026-04-17) — executive compensation (Kim $38.2M, Ma/Sanchez), board composition, controlled-company status, classified board, AE Industrial director-nomination rights, related-party agreements, say-on-pay exemption. https://www.sec.gov/Archives/edgar/data/1860160/000119312526161474/fly-20260417.htm
4. Form S-1 (follow-on) (filed 2026-05-26) — May-2026 offering; selling-securityholder table (AE Industrial as sole seller); use of proceeds; updated capitalization and risk factors. https://www.sec.gov/Archives/edgar/data/1860160/000119312526239249/d104612ds1.htm
5. 8-K — Follow-on pricing (filed 2026-06-02; event 2026-05-28) — offering priced $48.00; 4.0M primary + 8.0M secondary + 1.8M option; underwriters. https://www.sec.gov/Archives/edgar/data/1860160/000119312526252407/d156539d8k.htm
6. 8-K — Q1 2026 results (filed 2026-05-04). https://www.sec.gov/Archives/edgar/data/1860160/000186016026000008/fly-20260504.htm
7. 8-K — Q4/FY2025 results (filed 2026-03-19). https://www.sec.gov/Archives/edgar/data/1860160/000119312526115970/fly-20260319.htm
8. 8-K — SciTec acquisition completion (filed 2026-11-05; event 2025-10-31). https://www.sec.gov/Archives/edgar/data/1860160/000119312526… (Item 2.01)
9. 8-K — Reorganization / SciTec agreement (filed 2025-10-06; event 2025-10-05).
10. 8-K — Credit agreement amendments (2025-11-10 to $260M; 2026-04-03 to $305M).
11. 8-K — Alpha FLTA006 update (filed 2025-08-26) — FAA clearance to resume Alpha launches.
12. 8-K — IPO underwriting agreement (filed 2025-08-08; event 2025-08-06) — 22.2M shares at $45.
13. Forms 3/4/5 (2025-08 through 2026-06) — insider/AE Industrial transactions; the June-2026 AE secondary sale (8.0M @ $48); IPO-era grants; no code-P open-market purchases.

Primary — Company sources

14. Q1 2026 earnings call transcript (2026-05-04) — record revenue $80.9M; segment mix; FY26 guide $420–450M; Golden Dome/FORGE; NASA Moon Base “$20B/7yr” framing; Blue Ghost Mission 2 ($130M); Alpha Flight 7; Eclipse status; liquidity $811.6M. (Via ROIC.ai transcript service; company IR: investors.firefliespace.com.)
15. Firefly IR / company releases — Blue Ghost Mission 1 landing (2025-03-02); VICTUS NOX/DM; NVIDIA partnership; NASA CLPS $144M award (2026-06-30). https://www.fireflyspace.com

Quantitative data providers (cross-check; reconciled to filings)

16. ROIC.ai — income statement, balance sheet, cash flow, enterprise value, profitability/valuation ratios (FY2023–Q1’26). Note: ROIC’s SciTec purchase-price ($855M) and employee count (~780) conflicted with the 10-K ($547.0M; 1,409 employees); the filing figures were used.
17. AZI price history — 5-year (IPO-to-date) daily OHLCV CSV; used for the price-action event map, 52-week range ($16.00–$73.80), beta (~3.2). (AZI valuation-index own-history percentile returned null — insufficient history for a fresh IPO.)
18. AZI news feed — recent-events timeline (KeyBanc upgrade/$50 PT; SpaceX-IPO sector moves; Space-ng; NASA CLPS award).
19. FactorsToday — returned no data for FLY (IPO <1 year; <252-day history); factor-loading/leaderboard read unavailable, noted in the memo.

Peer / industry cross-reference (public filings)

20. Rocket Lab (NASDAQ: RKLB) — direct launch comp; Electron/Neutron; EV/sales, burn, backlog (per RKLB SEC filings, mid-2026).
21. Intuitive Machines (NASDAQ: LUNR) — direct lunar/CLPS comp; both IM landers tipped over; EV/sales; CLPS margin economics (per LUNR SEC filings, mid-2026).
22. AST SpaceMobile (NASDAQ: ASTS) — capital-intensive space “valuation-of-hope” comp (per ASTS SEC filings, mid-2026).
23. Redwire (NYSE: RDW) — space infrastructure/components; EAC-overrun precedent; dilution comp (per RDW SEC filings, mid-2026).

Third-party market data (as cited in Firefly filings)

24. McKinsey (2024) — $1.8T space economy by 2035; WEF/McKinsey — $32B launch market by 2035; Research and Markets — $7B spacecraft market (2024); BryceTech (2025) — ~2,800 satellites launched 2024. (Cited by the company; treated as promotional TAM framing, not independently verified.)

Notes on data quality

• SciTec price and headcount: the 10-K (Note 3) governs — $547.0M purchase price, 1,409 total employees — over ROIC’s aggregator figures.
• Backlog: the “$1.4B total backlog” is a management non-GAAP figure; audited remaining performance obligations were $684.9M.
• “Ocula”: referenced in the earnings call / IR as an announced service; not a disclosed revenue line in the 10-K/10-Q/S-1.
• Space-ng acquisition: reported in June-2026 news but not found in the local filings; flagged as an open item.