Snapshot

This sub-sector covers the companies that build, launch, and operate satellites for defense reconnaissance, missile warning, communications, and space domain awareness, plus those that analyze satellite imagery for intelligence customers. The unit of output is individual satellites (or constellation slots) and orbital launches. Government customers — the Space Development Agency (SDA), Space Force, National Geospatial-Intelligence Agency (NGA), and allied militaries — are the dominant buyers. Around 2022-2023, the Pentagon shifted from a handful of exquisite, multi-billion-dollar geostationary (GEO) satellites to hundreds of cheaper, mass-produced low-Earth-orbit (LEO) satellites refreshed every few years — the "proliferated" architecture. That shift created a manufacturing and launch volume problem the legacy primes were not built for, opening the door to smaller, faster builders like Rocket Lab and BlackSky.

The Pentagon is building a military internet in space: hundreds of networked LEO satellites for missile tracking, data relay, and eventually orbital interception. Over $9.6B has been awarded by SDA across Tranches 0-3 for 344 satellites, but only Tranche 0 (20 satellites) is fully on orbit; the rest are in manufacturing or early launch campaigns with timelines that have slipped repeatedly. The Golden Dome homeland-defense program adds a space interceptor layer with $3.2B in early-stage contracts across 12 companies.

Source: starlust.org 2025 launch statistics; SDA Tranche 3 award (DefenseScoop, Dec 2025); Golden Dome SBI awards (Satellite Today, Apr 2026).

The product & how money is made

Four distinct products, each with its own revenue model:

• Launch services — physically delivering satellites to orbit on a rocket. Revenue is per-mission: Rocket Lab charges ~$6-7.5M per Electron launch (small-lift, ~300 kg to LEO) and ~$9.5M per HASTE suborbital mission for DARPA. SpaceX's Falcon 9 rideshare starts at ~$1.1M per slot; dedicated government missions cost far more. Rocket Lab's Neutron (medium-lift, ~13,000 kg, reusable) targets $50-55M per mission est. with test flight scheduled for H2 2027.
• Satellite manufacturing — designing and building the spacecraft. Revenue is typically firm-fixed-price contracts paid on delivery milestones. Rocket Lab builds spacecraft and components (via acquisitions: SolAero solar panels, Sinclair star trackers, Planetary Systems separation systems). Lockheed Martin and Northrop Grumman build large, complex satellites including classified payloads. SDA Tranche 3 tracking per-satellite contract values range from ~$11M (Rocket Lab's $805M ÷ 18) est. to ~$61M (Lockheed's $1.1B ÷ 18) est..
• Satellite imagery & analytics — operating a constellation and selling the data or analysis. BlackSky operates 15+ satellites (4 Gen-3 at 35cm resolution) and sells imagery subscriptions plus AI-powered analysis to NGA, defense, and international intelligence customers. Revenue is recurring subscription or per-tasking fees. Planet Labs (PL, not in this sheet's ticker set) is the largest Earth-observation constellation at 200+ satellites.
• Satellite components and subsystems — Lockheed Martin and Northrop Grumman supply satellite buses, sensors, and propulsion to other satellite manufacturers. Northrop Grumman's GEM 63 solid rocket motors (used in ULA's Atlas V and Vulcan rockets) are a significant space-adjacent product line.

For LMT and NOC, "space" is one segment of a multi-segment defense conglomerate. Lockheed Martin Space generated $13.0B in 2025 (17% of LMT total revenue). Northrop Grumman Space Systems generated $10.8B in 2025 (26% of NOC total revenue, down 8% YoY due to classified program wind-downs). For RKLB and BKSY, space is the entire company.

Source: RKLB Q1-2026 earnings; howtheymake.money analysis; LMT Q4-2025 earnings (Via Satellite, Jan 2026); NOC Q4-2025 earnings (payloadspace.com, Jan 2026); SDA Tranche 3 awards (DefenseScoop, Dec 2025).

Demand

Contracted demand contracted

The SDA's Proliferated Warfighter Space Architecture (PWSA) is the single largest source of contracted satellite demand:

Total contracted PWSA satellites across Tranches 0-3: 344 satellites worth ~$9.6B. The constellation's ultimate target is 300-500+ transport-layer satellites est., with additional tracking and custody layers on top. Rocket Lab's $805M SDA Tranche 3 win is the company's first missile-warning satellite contract.

Company-level backlogs as of most recent filings:

RKLB's backlog grew 20.2% quarter-over-quarter to $2.2B, with government/defense at 49% of backlog ($1.08B), up from $0.65B a quarter earlier. Key new contracts: $190M DARPA MACH-TB 2.0 (20 HASTE launches over 5 years), plus Golden Dome SBI selection and Anduril partnership for Neutron.

BKSY raised 2026 guidance to $130-150M revenue (from $120-145M), with $160M in contracts awarded year-to-date including a $99M sole-source Air Force Research Lab contract for advanced optical payloads.

Source: SDA awards (sda.mil, DefenseScoop, Air & Space Forces); RKLB Q1-2026 earnings (stocktitan.net); BKSY Q1-2026 transcript (Motley Fool, May 2026); LMT Q1-2026 earnings (lockheedmartin.com, Apr 2026); NOC Q4-2025 earnings (Satellite Today, Jan 2026).

Forecast demand est.

Global orbital launch attempts grew 25% year-over-year (263 in 2024 to 329 in 2025), and satellites placed in orbit grew 58% (to 4,517). SpaceX dominated with 170 of 181 US launches (~94% US share), but the majority were Starlink commercial broadband, not defense. Rocket Lab flew 17 successful missions from New Zealand in 2025 out of 18 attempts.

Defense-specific demand drivers beyond existing SDA contracts: Space Force's FY2026 budget request included increased funding for proliferated LEO systems. The National Reconnaissance Office (NRO) is building its own proliferated architecture for intelligence imaging. BlackSky reported "significant international demand drove the majority of backlog growth" in 2025 and signed an "eight-figure multi-year contract with an international defense customer" for Gen-3 satellites and imagery.

Source: starlust.org 2025 launch stats; BKSY Q4-2025 earnings (ir.blacksky.com); RKLB Q1-2026 filings.

Supply

Launch capacity

Global supply of orbital launches reached 329 attempts in 2025, with SpaceX providing ~52% of all global attempts. The binding constraint is dedicated, responsive launch for national-security payloads — government satellites often require dedicated launches (no rideshare) on rockets with the right orbit capability and security clearances. Rocket Lab's Electron is the only operational US small-launch vehicle besides SpaceX, with 91 cumulative launches through Q1 2026. The company ran 6 missions in Q1 2026 (vs. 3/quarter a year prior) and booked a record 31 missions in the quarter including 20 HASTE missions.

Neutron (Rocket Lab's medium-lift, reusable rocket) first test article assembly is targeted for Q3 2026 est., with test flights in H2 2027 est. — but delays into 2028 are plausible given typical rocket development timelines. Neutron targets the constellation-deployment market (13,000 kg to LEO) currently served only by SpaceX Falcon 9.

Satellite manufacturing capacity

The proliferated LEO shift created a manufacturing bottleneck. Legacy primes (LMT, NOC) are organized around building a few large, expensive satellites per year on 3-5 year timelines. The SDA model requires dozens of smaller, cheaper satellites delivered in 18-24 month cycles. Rocket Lab has positioned for this via vertical integration — it acquired SolAero (solar cells), Sinclair (star trackers), Planetary Systems (separation systems), and most recently Mynaric (laser communication terminals, closed May 7, 2026). The company claims its spacecraft and components have enabled more than 1,700 missions to date.

LMT delivered 21 satellites for a Tranche 2 transport-layer launch and is building for multiple SDA tranches simultaneously. NOC's space revenue declined 8% in 2025, partly due to SDA satellite timing and classified program wind-downs, but space backlog grew 13% to $26.2B, largely on GEM 63 rocket motor orders (supporting Amazon's Project Kuiper constellation) and SDA Tranche 3.

Earth-observation capacity

BlackSky operates 15+ satellites (mix of Gen-2 and Gen-3), with 4 Gen-3 satellites delivering 35cm very-high-resolution imagery. The company targets "at least 8 Gen-3s on orbit" by end of 2026 est.. Planet Labs operates 200+ satellites. Maxar (now part of Advent International) operates the highest-resolution US commercial constellation. NGA has expanded commercial imagery procurement through the Electro-Optical Commercial Layer (EOCL) program.

Key bottlenecks

• Security-cleared space engineers — extremely scarce, limiting classified payload production
• Space-qualified components — radiation-hardened chips, optical inter-satellite link terminals, and reaction wheels have long lead times
• Launch manifest congestion — government customers compete with commercial customers for limited dedicated launch slots
• Satellite manufacturing throughput — the primes are tooled for dozens per year, not hundreds; new entrants (York Space, Rocket Lab) are building higher-rate production lines

Source: RKLB Q1-2026 earnings; NOC Q4-2025 earnings (Satellite Today, Jan 2026); BKSY Q1-2026 transcript; starlust.org 2025 launch data.

The gap

The SDA has contracted for 344 satellites across Tranches 0-3, with a target constellation of 300-500+ transport-layer satellites est. plus tracking, custody, and interceptor layers. Only Tranche 0 (20 satellites) is fully on orbit. Tranche 1 (126 satellites) began launching in September 2025. Tranches 2 and 3 (198 combined) are in manufacturing. The Golden Dome interceptor layer adds an undefined number of additional satellites in the late-2020s timeframe.

Global launch capacity grew 25% in 2025 but is dominated by one provider (SpaceX at 52% of global attempts, 94% of US attempts). For national-security customers who need launch diversity, available capacity is far thinner than headline numbers suggest. Rocket Lab is the primary alternative, but Electron is small-lift only (~300 kg). Neutron's arrival would add a second US medium-lift provider.

Rocket Lab's average selling price per mission is rising: HASTE missions for DARPA command ~$9.5M vs. ~$6-7.5M for commercial Electron. The SDA is paying $11-61M per satellite est. depending on complexity and vendor. BlackSky targets 80% gross margins on its subscription business est. and guided 2026 revenue growth of ~30% at the midpoint.

Source: SDA constellation targets (sda.mil); RKLB per-mission pricing (howtheymake.money); BKSY Q1-2026 transcript.

The players

What each player does differently

• RKLB (Rocket Lab) — vertically integrated: builds rockets (Electron, developing Neutron), satellite components, and complete satellites for SDA. Only non-SpaceX US launch provider at operational scale. Q1 2026: space systems revenue ($84.5M) surpassed launch services ($84.1M) for the first time. Defense backlog jumped from $0.65B to $1.08B in one quarter. Not yet profitable; expects operating profitability Q3-Q4 2026 est..
• LMT (Lockheed Martin) — the largest defense space contractor. Builds SDA transport satellites (Tranches 1-3), the Next Generation Interceptor (NGI), Fleet Ballistic Missiles, Orion spacecraft, and classified payloads. Won SDA Tranche 3 tracking at $1.1B (18 sats). Space is 17% of total company. Space operating margin compressed to 8.2% in Q1 2026 (from 11.8%) due to absence of prior-year favorable program completions.
• NOC (Northrop Grumman) — builds satellites, rocket motors (GEM 63 for ULA/Amazon), and operates classified space programs. Space is 26% of total company. Revenue declined 8% in 2025 ($10.8B) due to classified program wind-downs, but space backlog grew 13% to $26.2B. GEM 63 orders (supporting Amazon Project Kuiper) stretch into the 2030s est.. Won SDA Tranche 3 tracking at $784M (18 sats). Guides space revenue of ~$11B in 2026.
• BKSY (BlackSky) — satellite imagery and AI analytics. Operates 15+ satellites (4 Gen-3 at 35cm resolution). Sells intelligence subscriptions to NGA and international defense customers. FY2025 revenue $107M; Q1 2026 revenue $20.8M (TTM basis distorted by lumpy contract timing). Guided 2026 revenue $130-150M. 321 employees, 69% gross margins, -89% operating margin. Debt of $211M against $118M cash. Guided $12-24M adjusted EBITDA in 2026.

Source: yfinance live data (Jun 2026); RKLB Q1-2026 (stocktitan.net, howtheymake.money); LMT Q1-2026 (lockheedmartin.com); NOC Q4-2025 (Satellite Today, payloadspace.com); BKSY Q1-2026 transcript (Motley Fool) and Q4-2025 results (ir.blacksky.com).

The price of exposure

RKLB trades at 97.7x trailing revenue with no current earnings. Each $1 of market cap buys $0.01 of current revenue and $0.033 of backlog. Trailing revenue growth is 63.5% YoY and defense backlog grew 66% in one quarter. Neutron is not yet flying.

LMT and NOC generate current cash flow ($4.0B and $2.1B FCF respectively) with large backlogs, but only 17% and 26% of revenue respectively comes from space. A dollar of LMT market cap buys $0.33 of space backlog; the remaining $1.25 of total backlog is F-35s, missiles, and rotary aircraft. LMT's forward P/E is 16x; NOC's is 17.5x.

BKSY has a $1.4B market cap and trades at 14.2x trailing revenue. Gross margins are 69%, but operating expenses exceed total revenue (-89% operating margin). At current run rates, cash ($118M) and FCF burn (-$57M/year) imply roughly 2 years of runway est. before additional capital is needed. Debt ($211M) is nearly 2x cash. The company guides $12-24M adjusted EBITDA in 2026.

Source: yfinance live data (Jun 2026); backlog figures from respective company filings as cited above.

What to deep-dive next

• Neutron economics and timeline risk — Rocket Lab's valuation depends heavily on whether Neutron flies on schedule (H2 2027) and at what per-mission price. Test article assembly (Q3 2026) is the next concrete milestone. Rocket development programs historically slip 1-3 years.
• SDA delivery execution — Tranche 1 is launching now. Track whether the 126 satellites reach orbit on schedule and work. Is Rocket Lab's $805M Tranche 3 win at a fixed price that can generate profit, or is it a break-even credential-builder?
• Golden Dome down-select — $3.2B across 12 companies is early-stage development money. Production contracts will be larger. Which companies down-select into full production?
• BKSY cash and debt — Verify whether subscription revenue reaches the $100M run rate at 80% margins. Check the $211M debt maturity schedule.
• LMT Space margin compression — Space operating margin fell from 11.8% to 8.2% in Q1 2026. One-quarter issue (absence of prior favorable completions) or structural (fixed-price cost growth)?
• NOC classified program backfill — $10.8B space revenue declined 8% due to classified wind-downs. How quickly do new restricted programs backfill?
• SpaceX competitive positioning — SpaceX launches 94% of US missions and manufactures its own defense satellites (Starshield). It won Golden Dome SBI work. If SpaceX bundles launch+satellite at scale, what happens to standalone satellite manufacturers?

Sources & confidence