Eutelsat

Phase A — Understand the business

Lens 1 · Company Overview

Eutelsat Group is the world's first fully integrated GEO-LEO satellite operator, created by the September 2023 merger of France's Eutelsat Communications (a legacy geostationary operator founded 1977 as an intergovernmental org, privatised 2001) with Britain's OneWeb (a low-Earth-orbit broadband constellation that went through Chapter 11 in 2020 and was rescued by the UK government + Bharti). The combined entity operates ~35 GEO satellites plus a LEO constellation of 600+ satellites.

The business sells satellite capacity and connectivity across four operating verticals:

1. Video — broadcasting capacity leased to TV channels/platforms (the legacy cash cow, structurally declining). ~46% of operating-vertical revenue in H1 25-26.
2. Fixed Connectivity — broadband to enterprises, ISPs, telcos (increasingly LEO-delivered).
3. Mobile Connectivity — in-flight connectivity (IFC) for airlines + maritime; the fastest-growing structural end-market.
4. Government Services — sovereign/defence comms; the strategic core of the new thesis (Ukraine, French military, IRIS²).

Contract structure: a mix of multi-year capacity leases (Video — recurring but eroding) and newer connectivity service agreements (LEO — usage/seat-based, faster-growing). Backlog €3.4bn at 31 Dec 2025 vs €3.7bn a year earlier — ~2.7x trailing revenue, Connectivity now 59% of it. The declining backlog is itself a tell: the recurring Video annuity is running off faster than connectivity bookings replace it.

The real identity: this is not a clean growth story. It is a leveraged, ex-utility broadcast-capacity business in secular decline that has strapped on a capital-hungry, unprofitable LEO constellation and been re-cast — by geopolitics, not by its own P&L — as Europe's answer to Starlink. The equity now trades as a political/sovereign option rather than a cash-flow security (see Lens 8).

Lens 2 · Supply Chain

Map: manufacturers → Eutelsat (capacity owner/operator) → distributors/integrators → end users.

• Upstream — satellite manufacture:
• Airbus Defence and Space — prime contractor for OneWeb Gen2. Contracted Dec 2024 for 100 satellites, expanded Jan 2026 to 440 total new LEO satellites; first batches deliver from late 2026. Single-source chokepoint for the constellation refresh.
• OneWeb Satellites (the legacy Airbus–OneWeb JV in Florida) built the Gen1 fleet on a quasi-mass-production line — the original cost advantage of the OneWeb model.
• Thales Alenia Space — a partner on IRIS² and European GEO programs.
• Launch: historically Arianespace (Soyuz pre-Ukraine, then Ariane) and — awkwardly — SpaceX Falcon 9, which launched a tranche of OneWeb Gen1 after the Russian Soyuz route was severed in 2022. The company depends on its principal competitor for launch — a structural irony and a chokepoint.
• Ground segment: gateway/teleport network + user terminals. Terminal cost/availability has historically been a OneWeb bottleneck (enterprise/gov terminals are expensive vs Starlink's mass-produced consumer dish).
• Midstream — Eutelsat: owns/operates the orbital capacity.
• Downstream — distribution:
• Bharti Airtel (also the second-largest shareholder) — distribution into India/South Asia via a Hughes/Bharti JV.
• Telco/ISP partners, IFC integrators, and defence primes who package capacity for end customers.
• End customers: TV broadcasters (Video); airlines — 15 committed, 160+ private jets, Japan Airlines 40+ widebodies; maritime; governments — French military (10-yr deal up to €1bn), EU/Ukraine.

Chokepoints: (1) Airbus as single-source for Gen2 satellites; (2) launch dependence on SpaceX/Arianespace; (3) user-terminal cost/supply; (4) capital itself is the binding input — the constellation cannot be sustained without continuous fresh financing (Lens 5/11). Names present — this lens passes.

Lens 3 · Competitive Advantages (moats)

The honest verdict: the durable moat is regulatory/political, not economic.

• Orbital spectrum & landing rights — Eutelsat holds scarce, grandfathered GEO orbital slots and Ku/Ka spectrum priority, plus OneWeb's Ku-band LEO priority filings. Spectrum rights are a genuine regulatory moat — but a wasting one as the band fills with competitors.
• Sovereign / "non-American" status — the single most valuable asset post-2025. Eutelsat is the only at-scale, European-controlled, multi-orbit operator. For European governments who do not want to depend on Elon Musk's Starlink, there is no alternative. This is why the French state stepped in. It is a moat of availability, not of cost or technology.
• GEO-LEO integration — Eutelsat markets itself as the only operator able to bundle GEO (high-throughput, fixed beams) with LEO (low-latency, global) in one service — relevant for IFC and government, where resilience/multi-path matters. Real but narrow.
• Switching costs — moderate in Video (multi-year leases, channel neighbourhoods) and government (integration, security accreditation); low in commodity broadband.

Bargaining power: weak on both sides. Upstream, Eutelsat needs Airbus and SpaceX more than they need it. Downstream, against Starlink's price/scale it has little pricing power in commodity connectivity — which is precisely why management explicitly says it does not compete with Starlink in consumer and retreats to government/enterprise/IFC niches. A moat you retreat into is a weak moat. Ground in: no commercial-layer files exist for space (all missing per Step 0), so this is web-derived.

Lens 4 · Segments

segments.csv is empty — all figures ``, from H1 FY25-26 (six months to Dec 2025) and Q3 FY25-26 (Jan–Mar 2026) disclosures. Eutelsat's fiscal year ends 30 June.

The structural story in one line: Video (the cash) is melting at low-double-digits; the three connectivity verticals + LEO are compounding at 15–65% but off a small base and not yet profitable at the constellation level. Connectivity is now ~54% of operating-vertical revenue and LEO ~20% of group. The mix shift is favourable in direction but the absolute connectivity dollars are nowhere near large enough to fund the constellation they depend on — that is the entire problem (Lens 11).

---

Phase B — Measure performance

Lens 5 · Earnings Result

No quarterly P&L on the shelf. Latest hard markers, all ``:

FY 2024-25 (year to 30 Jun 2025):

• Revenue €1,244m, +2.5% reported / +1.6% like-for-like; four operating verticals €1,226m (+0.8% LFL).
• Adjusted EBITDA €676.2m, margin 54.2% (constant FX) — flat LFL.
• Net debt / Adjusted EBITDA 3.88x at 30 Jun 2025 (vs 3.79x a year earlier) — rising leverage, the proximate trigger for the recapitalisation.
• LEO revenue grew >80%, ~15% of group.
• Liquidity ~€1.07bn (cash + undrawn lines).

H1 FY25-26 (to 31 Dec 2025):

• Net debt €1,300.9m, down €1,325.7m vs Jun 2025 — almost entirely the €1,475.8m net capital-raise proceeds. This is the single most important number in the dossier: leverage fell from 3.88x toward ~1.9x because the state recapitalised it, not because the business de-levered organically.
• Weighted-average debt maturity 2.3 years (down from 3.0y) — a near-term refinancing wall.

Q3 FY25-26 (Jan–Mar 2026):

• Total Connectivity revenue €155.7m, +15.3%; LEO +65%.
• FY25-26 guidance confirmed: four verticals roughly flat vs FY24-25; LEO +50%; FY capex cut to ~€900m (from €1.0–1.1bn).

Margin/balance-sheet flags: 54% EBITDA margin is genuinely high (capacity businesses have low marginal cost) — but it is pre the LEO capex sinkhole. The gap between a 54% EBITDA margin and a business that cannot fund itself is the whole point: D&A + capex on the constellation consume the EBITDA. Discretionary free cash flow has been negative/marginal as capex runs ~€900m–€1.1bn against ~€676m EBITDA. Net income is not a clean positive figure (heavy D&A, impairments on legacy GEO) — n/a — not cleanly sourced for a single net-income line; do not fabricate one.

Market reaction: the equity has de-rated structurally despite "in-line" prints — the market is pricing the dilution and the funding gap, not the quarterly revenue (Lens 8).

Lens 6 · Earnings Calls (sentiment trend)

No transcripts/ on the shelf — drawn from CEO commentary and results coverage ``.

• Tone shift, 2024 → 2026: from a defensive integration / debt-management posture under Eva Berneke (CEO through May 2025, who delivered the OneWeb merger) to a sovereign-mission / multi-orbit-pivot posture under Jean-François Fallacher (CEO from June 2025).
• What management now emphasises: "European sovereignty," "multi-orbit," IRIS², OneWeb Gen2 roadmap, government demand, capex discipline (the €900m cut, the 30% Gen2 capex reduction).
• What they stopped saying: any pretence of competing with Starlink in the consumer mass market. The framing is now explicitly niche/sovereign, which is candid but is also an admission that the broad-market LEO land-grab is lost to SpaceX/Amazon.
• Recurring phrase: "secure operational continuity" of the constellation — telling, because it signals the priority is survival/refresh, not expansion/share-gain.

Sentiment read: cautiously constructive on strategy, defensive on economics. Management is competent and honest about the constraints; the constraints are just severe.

Lens 7 · Comps

Peer table — Eutelsat vs key satellite operators. No space peers in the research index (Eutelsat is the first space name covered), so peers and multiples are pulled fresh ``. Multiples that cannot be cleanly sourced are marked n/a (per provenance discipline — no fabricated multiples).

Eutelsat EV (for its own line) ``: market cap ~€2.6bn + net debt €1.3bn (Dec 2025) ≈ EV ~€3.9bn; EV/EBITDA ~5.7x. On EV/EBITDA Eutelsat looks "cheap" vs SES's ~9.3x — but the EBITDA multiple flatters a business whose EBITDA is consumed by capex, so it is the wrong lens. The right lens is FCF/equity-value, on which the name screens expensive (FCF marginal, equity a residual claim behind €1.3bn net debt + a €4bn forward capex commitment).

Read: the comp set splits into (a) profitable niche legacy (Iridium — the template Eutelsat wishes it were), (b) levered GEO incumbents (SES, Viasat — Eutelsat's actual peer group), and (c) pre-revenue LEO options (ASTS — pure speculation). Eutelsat is (b) wearing a (c) costume.

Lens 8 · Stock-Price Catalysts (the 5-year tape)

This is the most important lens for understanding what moves the name — because it has become a political instrument, not a cash-flow security.

• Mar 2025 — the +387% week. Trump paused US military aid to Ukraine and floated cutting Ukraine's Starlink access. Eutelsat — the only European at-scale alternative — soared: +77% on 4 Mar, +120% on 5 Mar, +387% over the week; market cap went from <$1bn to briefly >$4bn on speculation it could replace Starlink in Ukraine. Pure geopolitical re-rating, not a fundamental event.
• May 2025 — CEO change. Berneke out, Fallacher (ex-Orange) in; shares +10% on the news, read as French-state/telco consolidation of control.
• Jun 2025 — French state recapitalisation announced (€1.35bn, state to ~30%), one day after the French military's 10-yr OneWeb deal (up to €1bn).
• Nov–Dec 2025 — the dilution. €1.5bn capital raise completed: reserved tranches (~€828m, 21 Nov) + ~€670m rights issue (Dec). Share count 475m → 1,178m. 2.48x dilution.
• Dec 2025 — SoftBank selloff. Shares −7% on a report SoftBank was cutting its OneWeb-legacy stake.
• Feb 2026 — debt refi + state-backed satellite financing. €1.5bn senior bonds (2031/2033) to redeem 2027/2029 notes; €975m French-export-credit-backed financing for the 340 Airbus satellites.

Pattern: the stock reacts to geopolitics (Ukraine/Starlink/sovereignty) and capital-structure events (raises, refis, state backstops) — almost never to operational beats/misses. Q3 25-26 was "in line" and barely moved. Anyone owning this equity is making a bet on European political will and the French state's continued backstop, not on satellite unit economics. That is the single sharpest finding in this dossier.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO — Jean-François Fallacher (since Jun 2025). 30+ yrs European telecom; ex-CEO Orange France/Spain/Poland/Romania; ran nationwide fibre/5G rollouts and large public-private integrations. Track record: a telco operator/integrator, not a satellite or capital-markets specialist — fitting for a business that is now effectively a state-aligned infrastructure utility. Skin in the game: professional manager, not a founder; insider ownership negligible (insider-transactions.csv not present — `` only). The signal is that the French state, via APE, is the real principal, not the CEO.
• Predecessor — Eva Berneke (2022–May 2025) delivered the OneWeb merger, then was replaced as the state took control — a de facto governance reset.
• Capital-allocation history: poor on a returns basis. The OneWeb merger doubled the company into an unprofitable, capital-hungry constellation; leverage rose to 3.88x; the dividend was suspended; equity holders were diluted 2.5x to survive. ROE/ROIC have been low-to-negative through the build. This is value-preservation under state sponsorship, not value creation.
• Red flags (governance): heavy related-party density — Bharti Airtel is both ~18% shareholder AND the India distribution partner; the French state is controlling shareholder AND a major customer (military) AND the regulator's sponsor. These are not fraud flags but they mean minority equity holders' interests are subordinate to French/European industrial-policy objectives. A minority shareholder should assume decisions will be made for European sovereignty first, share price second.
• Archetype: state-backed national champion run by a professional telco operator. Implication: low bankruptcy risk (the state will not let it fail), low upside torque (the state will dilute/restructure rather than let equity holders capture a windfall).

Lens 10 · Forensic Red Flags

financials.csv is empty; assessment is `` + structural reasoning.

• Cash flow vs earnings: the central forensic feature is the chronic gap between Adjusted EBITDA (~€676m, 54% margin) and free cash flow (marginal/negative) — capex ~€900m–€1.1bn swamps EBITDA. "Adjusted EBITDA" is the metric management leads with precisely because the GAAP picture (post-D&A, post-capex, post-impairment) is far weaker. Trust FCF and net debt, not Adjusted EBITDA, on this name.
• Goodwill/intangibles: the OneWeb merger and legacy GEO fleet carry large intangible/goodwill balances exposed to impairment if LEO economics disappoint — a watch item (impairments have featured in prior years).
• Net debt definition & maturity wall: weighted-avg maturity just 2.3 years — the company is in continuous refinancing mode; the Feb-2026 bond deal addressed 2027/2029 notes but the structure remains short-dated.
• Off-balance-sheet / commitments: €4bn+ of forward capex commitments (OneWeb Gen2 €2–2.2bn + IRIS² €2bn) only partly financed — the real "liability" is contractual capex, not just reported debt.
• Dilution as a recurring instrument: 2.5x share issuance in one year. Future equity holders should price further dilution risk as the base case if the funding gap widens.

Regulatory findings (required sub-section):

• SEC (EDGAR LR + AAER): none possible — Eutelsat has no CIK and does not file with the SEC.
• Non-SEC enforcement (web search "Eutelsat" (FTC OR DOJ OR EU OR sanctions OR consent decree OR fine OR penalty)): No material fines or consent decrees surfaced. The one regulatory-adjacent item is EU/international sanctions on Russian broadcasters, which forced Eutelsat to drop Russian channels — this is Eutelsat complying with sanctions (a revenue headwind in Video, −12.3% partly attributed to it ), not an enforcement action against Eutelsat.
• Item 3 (Legal Proceedings): n/a — no 10-K on the shelf (foreign filer). French URD (Document d'Enregistrement Universel) would hold the equivalent; not ingested.
• Net: No material regulatory or legal enforcement findings against the company — verified via SEC EDGAR EFTS (no CIK, zero findings per the regulatory file) and web search as of 2026-06-30. The salient regulatory exposure is sovereign/political: as a French national champion its fortunes are tied to EU industrial policy and export-credit/state-aid rules.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

Eutelsat's fiscal year ends 30 June; projection covers FY2026, FY2027, FY2028. EPS is the wrong primary metric here (net income is not cleanly positive and dilution is the dominant variable) — so I anchor on revenue, Adjusted EBITDA, capex, and the funding gap, and give an indicative EPS only with heavy caveats. All with arithmetic; inputs.

Base case

• Revenue: FY26 ~€1.24bn (verticals flat, LEO +50% per guidance ) → FY27 ~€1.30bn → FY28 ~€1.38bn.
• Adjusted EBITDA: ~€670–700m/yr, margin drifting toward ~50% as mix shifts to lower-margin connectivity.
• Capex: ~€900m (FY26 guided ) → ~€800m–€1.0bn (FY27–28, Gen2 build).
• FCF: negative-to-breakeven through FY28 — the constellation refresh consumes EBITDA.
• EPS: ~breakeven to marginally positive on ~1.18bn shares; n/a — not reliably sourced as a point estimate, and not meaningful given dilution risk. Do not fabricate an EPS line.

Bull case: LEO compounds 50%+ for 3 years, government/IFC bookings accelerate, IRIS² and export-credit financing close the funding gap without further equity dilution, leverage holds ~2x, and a strategic/sovereign premium re-rates the equity. Revenue ~€1.5bn FY28, EBITDA margin holds ~52%, FCF turns positive FY28.

Bear case: Video declines accelerate (−15%+), Starlink/Amazon price-compress connectivity, the €4bn capex bill forces another equity raise (further dilution), leverage re-rises, and the state restructures over minority holders. Revenue stalls ~€1.2bn, FCF stays negative, equity de-rates toward distressed-asset value.

The decisive question is not EPS — it is the funding gap. Committed forward capex €4bn+ vs ~€676m annual EBITDA and €1.3bn net debt + ~€1.07bn liquidity + €975m export-credit facility. The base case requires near-flawless execution on financing (export credit + IRIS² EU money) to avoid a fourth capital event. That is the call.

(Forecast tracker: forecast.ts create is skipped per --watchlist rules — breadth mode logs no Brier forecast. Were one logged, the scoreable claim would be a binary: "Eutelsat completes OneWeb Gen2 funding through FY2028 without a further equity raise," p≈0.40.)

Lens 12 · Bull vs Bear

Bull case. Eutelsat is the only investable, at-scale, European-sovereign multi-orbit operator in a world that has decided — post-Ukraine — that depending on Elon Musk's Starlink is a strategic vulnerability. It has the orbital spectrum, a 600+ LEO constellation already in service, a GEO cash base, a 10-year French military anchor (up to €1bn), a seat in IRIS² (Europe's €10.6bn sovereign constellation), and a French state that has demonstrated it will write the cheque. LEO revenue is compounding 50–65%. If Europe's defence/sovereignty capex super-cycle is real, Eutelsat is its listed pure-play, with the state de-risking the downside. The EV/EBITDA (~5.7x) is a fraction of growth-LEO peers.

Bear case (the three permanent-impairment risks).

1. The funding gap is structural, not bridgeable on current cash flow. €4bn+ committed capex against ~€676m EBITDA and negative FCF means the equity is a perpetual call on dilution. The 2.5x dilution of 2025 is the template, not the exception.
2. The product is competitively cornered. Management has conceded the consumer mass market to Starlink and is retreating to government/enterprise niches — exactly where Amazon Leo ($10bn committed, enterprise beta Apr 2026 ) and Starlink's gov tier also compete. The niche is not defensible enough to earn a return on €4bn of capex.
3. Minority equity is structurally subordinated. With the French state at ~30% and controlling, decisions will be made for European industrial policy — equity holders are residual claimants behind the state's strategic agenda and €1.3bn of net debt.

Pre-mortem (18 months out, thesis broke): Video declined faster than modelled, LEO connectivity ARPU compressed under Amazon/Starlink, the export-credit financing came with strings or fell short, and Eutelsat launched another rights issue in 2027 — diluting holders again — while the share price ground toward €1.50. The "sovereignty premium" proved to be a one-time 2025 spike, not a durable re-rating.

Are multiples too high? On EV/EBITDA, no (it's optically cheap). On equity value vs free cash flow and dilution risk, yes — the equity is priced as an option, and options can expire worthless.

Contrarian view (what the market refuses to see): the market still half-prices the March-2025 fantasy that Eutelsat could "be Europe's Starlink." It can't — it's a niche sovereign carrier with a melting legacy book and a capex bill it can't self-fund. But the market also under-appreciates the flip side: the French state will not let it fail. So the real trade is not the equity (leveraged, dilutive) but the credit / state-backstopped instruments — the sovereign put is the asset, and it sits in the bonds, not the shares.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case:

• Revenue concentration & quality: 46% of revenue is structurally declining Video; the "growth" is ~20% LEO that doesn't cover its own capex. The backlog is shrinking (€3.7bn → €3.4bn) — the recurring annuity is running off faster than new bookings replace it.
• The moat is rented, not owned. Eutelsat depends on Airbus (sole-source satellites) and SpaceX (launch — its own competitor). It has no cost advantage vs Starlink's vertically integrated launch+manufacture. The only true moat — "European-controlled" — is a regulatory designation, not an economic one, and could be matched by any state-backed European new-entrant.
• Most dangerous competitor bulls underestimate: not Starlink (priced in) but Amazon Leo — $10bn committed, 1 Gbps enterprise beta (Apr 2026 ) — coming straight for Eutelsat's enterprise/gov niche with Amazon's balance sheet and AWS distribution. And SES-Intelsat (post-2025 merger) as the consolidated European GEO peer.
• Worst capital allocation: doubling into OneWeb (an asset that had already been through Chapter 11), then diluting equity 2.5x to fund it. Related-party density (Bharti = shareholder + distributor; state = shareholder + customer + sponsor) means capital decisions aren't made for minority holders.
• Assumptions that must hold for today's price: (1) no further equity raise; (2) LEO 50%+ growth sustained; (3) export-credit/IRIS² financing closes the €4bn gap; (4) the sovereignty premium persists. If any one fails, the equity re-rates down.
• −20–30% growth scenario: if LEO growth halves and Video accelerates down, FCF stays negative indefinitely → a 2027 rights issue → another 30–50% dilution → equity toward distressed-asset value (€1.0–1.5/share ).
• Single permanent-impairment scenario, plausibility: a fourth capital event on worse terms (dilutive convertible or state-led restructuring) that wipes out most of the current equity value while the state/creditors are protected. Plausibility: moderate-to-high given the funding math — this is the base bear case, not a tail.

Lens 14 · Management Questions (ordered by information value)

1. What is the fully-financed funding plan for the €4bn+ OneWeb Gen2 + IRIS² capex through 2031, and what specifically prevents another equity raise? (The whole thesis turns on this.)
2. At what LEO revenue scale does the constellation generate positive free cash flow, and in which fiscal year do you model group FCF turning sustainably positive?
3. Given you've conceded the consumer mass market to Starlink, what is the defensible long-run market share and ARPU in your government/enterprise/IFC niches once Amazon Leo is at scale?
4. How should minority equity holders think about the alignment of the French state (controlling shareholder + customer + policy sponsor) with their interests when the two conflict?
5. What is the expected dilution path under your base/bear cases — should equity holders price further issuance as a base case?
6. With weighted-average debt maturity at ~2.3 years, what is the refinancing roadmap, and at what cost of debt?
7. What is the realistic incremental margin of connectivity/LEO revenue vs the legacy 54% Video EBITDA margin — i.e. where does group margin settle?
8. How much of the €3.4bn backlog is Video (running off) vs Connectivity (growing), and what is the net new-bookings trajectory?
9. What are the terminal-cost economics for OneWeb user terminals vs Starlink, and how does that gate enterprise/gov adoption?
10. What is the return on invested capital you target on OneWeb Gen2, and how does it compare to the legacy GEO book?
11. How exposed is the IRIS² timeline (service early-2030s) to EU budget/political risk, and what is your contingency if it slips?
12. What is the strategic logic for remaining a standalone public equity rather than being taken fully private/sovereign by the French state?
13. What is the impairment risk on OneWeb goodwill/intangibles and the legacy GEO fleet under your bear scenario?
14. How do you assess launch dependence on SpaceX, and what is the path to non-competitor launch capacity (Ariane / others)?
15. What single metric should investors watch each quarter to know whether the LEO economics are working (not just LEO revenue growth)?

---