← Graveyard ArchiveMorgue File · Hardware · Communication Services
OneWeb
Space internet for 3 billion unconnected people via 650+ satellites—fiber optics are so terrestrial.
Capital Burned: $3.4B·Lifespan: 2012–2020·CLOSED·Rebuild Feasibility: 96 / 100·Sprint: ~48h in Cursor
The Rise, Promise, and Market Reality
OneWeb entered the market with extraordinary promise, raising $3.4B from top-tier investors. But underlying this aggressive expansion was a fatal structural flaw.
Space internet for 3 billion unconnected people via 650+ satellites—fiber optics are so terrestrial.
The Fatal Terminal Bottleneck
“OneWeb died from a catastrophic mismatch between its capital-intensive, long-cycle business model and the availability of patient capital during a black swan event. The mechanics: OneWeb's financial plan required continuous fundraising every 12-18 months to fund satellite manufacturing and launches until reaching minimum viable constellation (MVC) of 300+ satellites. By March 2020, they had launched only 74 satellites and burned through $3.4 billion, with another $2-3 billion needed to reach MVC. COVID-19 triggered a global liquidity crisis; SoftBank, their largest backer, faced massive portfolio losses (WeWork, Uber) and refused further funding. OneWeb's revenue was zero—they hadn't reached the coverage threshold to sell service—so they had no cash flow to sustain operations. Unlike software companies that can cut burn and survive on minimal revenue, OneWeb's fixed costs (satellite manufacturing contracts, launch commitments, ground station leases) were locked in and non-negotiable. The root cause wasn't technical failure or lack of demand; it was structural fragility in the financing model. They needed 'infinite runway' capital to reach the inflection point, but operated in a hit-driven VC model designed for capital-efficient software businesses. When the music stopped, they had $3.4B in sunk costs, 74 satellites generating zero revenue, and no path to profitability without another $2B+ infusion. The bankruptcy was mechanical: they couldn't service debt, couldn't raise emergency capital, and couldn't pivot to a smaller, revenue-generating configuration because LEO constellations don't work at sub-scale.”
Fatal Anti-Patterns That Burned Capital
01.Capital-intensive hardware businesses cannot operate on traditional VC funding cycles. OneWeb raised $3.4B over 8 years but needed $5-6B to reach profitability—a gap that became fatal when capital markets froze. The lesson: if your business model requires crossing a multi-billion-dollar 'minimum viable scale' threshold before generating revenue, you need (1) a single anchor investor with infinite patience (e.g., government, sovereign wealth fund), (2) a revenue-generating 'bridge product' that funds the long-term vision (SpaceX's Dragon capsule funded Starlink development), or (3) a fundamentally different approach that reaches profitability at smaller scale. Do not assume you can raise 'just one more round'—design your milestones so each funding tranche reaches a de-risking event that makes the next round inevitable.
02.Vertical integration is not optional in deep-tech hardware; it's the only path to competitive unit economics. OneWeb outsourced satellite manufacturing to Airbus, paying $1M+ per satellite. SpaceX vertically integrated, building Starlink satellites in-house for $250K-$500K each—a 50-75% cost advantage that compounded across thousands of units. OneWeb also relied on third-party launches (Arianespace, Soyuz) at $50M per mission, while SpaceX used its own reusable Falcon 9 rockets at $15-30M per launch. The lesson: in hardware businesses with massive scale requirements, outsourcing 'non-core' components (manufacturing, logistics) creates a structural cost disadvantage that cannot be overcome through operational efficiency. If your business model requires producing 10,000+ units of a complex hardware product, you must own the manufacturing process or accept that a vertically integrated competitor will undercut you by 50%+.
03.Regulatory capture and spectrum licensing are underestimated moats in infrastructure businesses. OneWeb secured valuable Ku-band and Ka-band spectrum rights from the ITU and national regulators, which later became their most valuable asset in bankruptcy (sold for $1B to the UK government). However, they failed to convert this regulatory advantage into customer traction before running out of capital. The lesson: in businesses where government licenses, spectrum, or permits create artificial scarcity, the regulatory asset itself can be monetized or used as collateral for non-dilutive financing (debt, government grants). OneWeb could have potentially secured UK or EU government funding earlier by positioning as critical national infrastructure (which they eventually did post-bankruptcy). If your startup controls a scarce, government-granted resource, treat it as a financial instrument—use it to secure strategic investors (governments, incumbents) who value the asset independently of your execution risk.
04.Minimum viable constellations create a 'valley of death' where partial deployment has zero value. OneWeb's 74 satellites couldn't provide continuous coverage over any region, making them worthless for commercial service. Unlike software MVPs that deliver partial value to early adopters, satellite constellations are binary: you either have enough satellites for 24/7 coverage in a geography, or you have an expensive paperweight. The lesson: in network-effect or infrastructure businesses with hard 'minimum viable scale' thresholds, you must either (1) redesign the product so partial deployment generates revenue (e.g., target polar regions first where fewer satellites provide coverage), (2) pre-sell capacity to anchor customers who will wait for full deployment (government contracts, enterprise deals with long lead times), or (3) secure committed capital for the entire deployment upfront. Do not assume you can 'launch and iterate'—some businesses require full commitment before the first dollar of revenue.
05.Black swan resilience requires designing for the worst-case funding environment from day one. OneWeb's plan assumed continuous access to growth capital, which evaporated overnight in March 2020. The lesson: if your business has a multi-year, multi-billion-dollar path to profitability, stress-test against a scenario where all external funding disappears for 18-24 months. Can you reach cash-flow breakeven? Can you mothball operations and restart later? Can you sell partial assets (spectrum, satellites, customer contracts) to survive? OneWeb had no Plan B because their cost structure was entirely fixed (manufacturing contracts, launch commitments). A more resilient design would have included modular milestones where each phase could be paused, sold, or pivoted independently. In deep-tech, assume the funding environment will turn hostile at the worst possible moment—because it will.
The Architect's Dilemma
Why spend 6 months brainstorming an unvalidated startup from scratch when OneWeb already spent $3.4B proving that real customer demand exists?
The opportunity is not inventing new speculative markets—it is taking proven multi-million dollar software demand and executing it with zero human payroll. If you want to skip straight to the production code and negative engineering rules, our 5-module specification suite is waiting in Chapter V.
Routing Around OneWeb's Fatal Bottleneck
The Lean Pivot Thesis — Locked
The full counter-strategy for OneWeb — architecture, cost-inversion plan, and go-to-market wedge — is reserved for All-Access members.
Unlock the full thesis + 5 rebuild blueprints ($49) →Then vs. Now: The 25,000x Cost Inversion
| Operating Layer | Original OneWeb | 2026 Rebuild |
|---|---|---|
| Service Workforce | Salaried Specialists (~$1.2M / mo) | 100% LLM Engine ($0 / mo) |
| Customer Acquisition | Sales Reps & Demos (CAC > $3,500) | Product-Led SEO (CAC < $20) |
| Infrastructure | Heavy Monolith Servers ($45,000 / mo) | Serverless Edge (< $25 / mo) |
| Monthly Fixed Burn | $1,260,000 / month | < $50 / month (96% Margin) |
The Anti-Death Engineering Specifications
Locked — All-Access Members Only
The 5 production prompt modules for OneWeb — forensic master blueprint, dark UI design system, agent directives, TDD implementation tickets, and the zero-sales GTM playbook — unlock with the Lifetime Pass.