SES

Phase A — Understand the business

Lens 1 · Company Overview

SES S.A. owns and operates a fleet of ~120 satellites across two orbits — geostationary (GEO, ~36,000 km) and medium-Earth orbit (MEO, ~8,000 km) — after completing the $3.1bn (€2.6–2.7bn cash + C-band CVRs) acquisition of Intelsat on 17 July 2025. It sells satellite capacity and increasingly managed connectivity services to three buyer types:

• Media (37% of FY25 revenue, ~€977m reported): broadcasters and content distributors. SES carries TV to ~2.3bn viewers. This is the legacy cash cow — high-margin, contracted, and structurally declining.
• Networks (62% of FY25 revenue): the growth engine, split into Government (€726m reported — largest and fastest-growing vertical), Aviation/IFC (€507m), and Fixed Data & Maritime (~€526m).

Reported FY2025: revenue €2,627m (+33.9% YoY reported), Adjusted EBITDA €1,196m (45.4% margin), Adjusted FCF €229m, net loss €95m. The reported growth is almost entirely the Intelsat consolidation (only ~5.5 months of it); the honest organic number is like-for-like revenue −1.6% and like-for-like Adjusted EBITDA −12.1%. That single line is the whole investment debate.

Contract structure / key terms. Capacity is sold on multi-year take-or-pay-style contracts, which is why SES carries a gross backlog of €6.6bn at 31-Dec-2025 (Media €3.0bn, Networks €3.6bn) and the combined group cites >€8bn including Intelsat's book. Backlog is the bull's best friend — it makes near-term revenue visible — and the bear's trap, because it masks the price at which capacity renews (renewals are repricing down; see Lens 3).

Lens 2 · Supply Chain

Map: spectrum + orbital slots + satellites → SES capacity/managed service → broadcasters, telcos, airlines, cruise lines, governments. Named stakeholders along the chain:

• Upstream — satellite manufacturers (chokepoint #1): Boeing (built the troubled O3b mPOWER MEO satellites — see Lens 10), Thales Alenia Space, Airbus Defence & Space, Lockheed Martin. Build cycles run 2–4 years and a single program slip directly defers revenue. SES is now running a board-level "CapEx taskforce" precisely because manufacturing/launch execution is its biggest controllable risk.
• Upstream — launch (chokepoint #2): SpaceX (the dominant launch provider — note the irony: SES pays its largest competitor, Starlink's parent, to put SES satellites up), Arianespace/Ariane 6, ULA. Launch scarcity and cost are structural.
• Upstream — spectrum/regulatory inputs: FCC (US C-band — source of the Intelsat C-band clearing windfall), ITU (global orbital-slot/frequency coordination), national regulators. Orbital slots + spectrum rights are the irreplaceable raw material and a genuine barrier (Lens 3).
• The company: SES operates the fleet + a global ground network (gateways, teleports) + increasingly a software/orchestration layer ("meoSphere", network-as-a-service).
• Downstream — customers: US DoD / Space Force / Army (largest single vertical), broadcasters, airlines (Intelsat brought a large in-flight-connectivity base, ~3,000 tails), cruise lines (Royal Caribbean-class operators via SES Cruise mPOWERED), telcos (cellular backhaul, and now Lynk/Omnispace for direct-to-device relay).

Single-source dependency to flag: SES depends on Boeing for the MEO constellation that is its differentiated growth asset, and the first-generation O3b mPOWER satellites shipped with a power-module defect (Lens 10). Manufacturing concentration on a troubled supplier is the supply-chain risk that already cost ~5% of 2024 revenue/EBITDA.

Lens 3 · Competitive Advantages (moats)

The moat is real but narrowing. Four sources, ranked by durability:

1. Spectrum + orbital slots (strongest, durable). GEO slots and C/Ku/Ka-band rights are finite, regulator-granted, and effectively un-buildable by a new entrant. This is why incumbents like SES and Intelsat had value worth combining. The C-band holdings even throw off windfalls (Intelsat earned $3.67bn of FCC accelerated-relocation payments; the FCC is exploring more C-band clearing, a potential future payout — captured partly via the CVRs SES issued).
2. Multi-orbit + government accreditation (the differentiated, defensible moat). SES is one of very few operators that can offer GEO + MEO + (via partners) LEO in a single managed service, with the security clearances, sovereign-Europe positioning, and integration that Western governments require. Starlink is fast and cheap but is a single-orbit, single-vendor, US-controlled commercial network — a non-starter for a sovereign-resilience mandate. This is the bull case in one sentence, and it is why Government grew +17.3% LFL while the rest shrank.
3. MEO latency/capacity guarantee (niche-durable). O3b mPOWER MEO delivers low-latency, guaranteed multi-gigabit capacity to a fixed site — better than LEO for mission-critical, capacity-assured use cases (cruise passenger Wi-Fi, government, enterprise). SES's own framing — "on a cruise ship, MEO runs the paid passenger Wi-Fi while Starlink lets crew call home" — is honest about where it wins and where it doesn't.
4. Switching costs (modest). Multi-year contracts + installed ground equipment create stickiness, but it is eroding as customers run dual networks (SES + Starlink) and use that optionality to push price.

Bargaining power — who needs whom? Against suppliers (Boeing, SpaceX) SES is weak (it needs them more than they need it). Against customers, power is deteriorating in data/media (oversupply + Starlink hand customers the leverage) but holding in government (where accreditation, sovereignty, and assured capacity make SES hard to displace). The moat, in short, is migrating from "we own the capacity" to "we are the trusted multi-orbit integrator for governments and mobility." Whether that migration outruns the decline of the legacy book is the entire thesis.

Lens 4 · Segments

Reported figures (EUR, FY2025, Intelsat consolidated only from 17-Jul-2025 — so reported growth massively overstates organic; the LFL column is the truth):

Trend and cause: the barbell is stark — Government + Aviation accelerating (sovereignty + connected-aircraft secular demand) against Media + Fixed Data decelerating hard (cord-cutting/SD shutdown in video; capacity oversupply + Starlink in data). The combined group's LFL Adjusted EBITDA fell from €1,783m (FY24) to €1,529m (FY25), −12.1% — i.e. the merged entity is currently less profitable than the two companies were separately. Synergies (€370m run-rate target) are the bridge meant to reverse that; until they land in the P&L, the segment math says the business is shrinking.

---

Phase B — Measure performance

Lens 5 · Earnings Result (FY2025, reported 2026-03-02)

The print. Revenue €2,627m (in the €2.60–2.70bn guide range); Adjusted EBITDA €1,196m at 45.4% margin (in the €1.17–1.21bn guide range); Adjusted FCF €229m (positive again); net loss €95m attributable to owners; net cash absorbed by investing (capex) €559m.

• Consensus beat/miss: No clean, separately-sourced consensus figure for SESG.PA was retrievable (search returns are contaminated by an unrelated "SES" battery company) — n/a. Against SES's own guidance, the print landed mid-range on revenue and EBITDA. Shares reportedly moved only modestly post-print (one outlet "+4% on EBITDA beat," another "−4% on the widened net loss") — i.e. the result was roughly in line and the stock is range-bound.
• What drove it: Government (+17.3% LFL) and Aviation (+28.5% LFL) carried the quarter; Fixed & Maritime (−14.6% LFL) and Media (−12.6% LFL) dragged. Net: LFL revenue −1.6%, LFL EBITDA −12.1%.
• Margin move: Adjusted EBITDA margin 45.4%, but LFL EBITDA fell more than revenue → negative operating leverage as the higher-margin media/data capacity reprices down and lower-margin managed-service/government revenue mixes in. PPA from the acquisition shaved €6m off revenue and €8m off EBITDA.
• Guidance/tone: 2026 guided "stable" revenue (€3.5bn LFL) and "stable" EBITDA (€1.5bn LFL) at constant FX — management is explicitly not guiding organic growth in 2026; the story is "hold the line, deliver synergies, deleverage." Capex guided ~€700m (incl. IRIS2 + first-phase meoSphere), ~€100m below prior guidance.
• Balance-sheet flags (the headline risk): Adjusted Net Debt €6,029m; net leverage jumped from ~1.1x to 3.9x post-Intelsat. Cash €674m (excl. €401m restricted for IRIS2). ~82% of debt fixed, ~4% weighted cost, ~5-yr weighted maturity. SES repaid ~€2,906m of maturities in 2025 (incl. $3bn of Intelsat's 6.50% 2030 notes). Plus three hybrid notes (€525m PerpNC26, €500m NC29, €500m NC32) that sit between debt and equity.
• Unusual vs. its own history: SES went from a low-levered, dividend-paying incumbent to a 3.9x-levered, loss-making integrator in one transaction. The dividend was cut to a €0.50/A-share base with a stated intent to raise it only after leverage falls below 3.0x (targeted 12–18 months post-close).

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on disk; sentiment reconstructed from press releases and CEO interviews across H1-25 → Q3-25 → FY25.

What management keeps saying (rising emphasis): "multi-orbit," "synergies / €370m run-rate," "government growth," "full-service space solutions," "disciplined capital allocation / deleverage to 3.0x," "meoSphere / innovation." CEO Adel Al-Saleh has consistently reframed SES from "satellite operator" to "full-service space solutions provider" and leans hard on the message that LEO doesn't solve every problem.

What they've started downplaying / stopped emphasizing: standalone media growth (now openly called "structural headwinds"); the O3b mPOWER insurance claim (a year-long, mostly-unpaid dispute they'd rather not foreground — Lens 10). The tone shifted across 2025 from "transformational deal" euphoria (H1) to "integration discipline and deleverage" sobriety (FY25) — appropriate, but it signals the easy narrative is behind them and the hard execution (synergies, leverage) is ahead.

Sentiment read: cautiously constructive on government/mobility; defensive on media/data; disciplined-to-the-point-of-conservative on guidance ("stable," not "growth"). The new board-level CapEx taskforce is a tell that the Board itself wants tighter control after the O3b mPOWER capex/anomaly experience.

Lens 7 · Comps

Western fixed-satellite + multi-orbit peers. Multiples are `` with date, or n/a. No multiple is fabricated. USD/EUR mixed as sourced; SES & Eutelsat in EUR.

Read. On the only two cleanly-sourced multiples, SES (7.40x EV/EBITDA) trades at a premium to Eutelsat (6.21x) — defensible given SES's stronger government franchise, MEO assets, and investment-grade intent, but both are deeply de-rated vs. the ~9–12x EV/EBITDA these names commanded pre-Starlink. The market is pricing the whole GEO/MEO complex as melting-ice-cube infrastructure: high cash flow, declining terminal value. The premium to Eutelsat is the market saying "SES is the better house in a bad neighbourhood," not "SES is cheap." The most important comp isn't in the table: Starlink/SpaceX (private, ~$350bn+ private mark), which is the demand-destruction engine repricing every public multiple here.

Lens 8 · Stock-Price Catalysts (what moves SESG.PA)

Pattern over recent years:

• The Intelsat deal (announced Apr-2024, closed Jul-2025) — the dominant catalyst of the period. Strategically cheered; financially it doubled leverage (1.1x→3.9x) and turned the group loss-making, capping the stock. Net reaction: muted-to-mixed, not a re-rating.
• C-band proceeds — Intelsat's $3.67bn FCC accelerated-relocation payment and any future C-band clearing (FCC actively exploring) are episodic positive catalysts; the CVRs SES issued tie shareholders to that optionality.
• O3b mPOWER anomaly (2023–) + insurance dispute — the Boeing power-module defect cut ~5% of 2024 rev/EBITDA and the ~$472m insurance claim remains largely unpaid — a recurring overhang.
• Dividend policy — the cut to a €0.50 base (raise contingent on <3.0x leverage) is a structural reset that capped the stock's appeal to the income holders who historically owned it.
• Starlink / LEO headlines — the secular de-rater. Every Starlink expansion (enterprise, maritime, direct-to-cell) compresses sentiment on GEO/MEO incumbents.

What it reveals: SESG.PA reacts most to (1) leverage/deleverage milestones, (2) C-band cash events, (3) the Starlink competitive narrative, and (4) dividend signals — more than to quarterly revenue beats. This is a balance-sheet-and-narrative stock, not an earnings-momentum stock. The re-rating trigger, if it comes, is "leverage through 3.0x + LFL EBITDA inflects positive", which unlocks the dividend raise and the "we survived Starlink" narrative simultaneously.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO: Adel Al-Saleh (since Feb 2024). Came from Deutsche Telekom, where he ran T-Systems (B2B IT services) — i.e. a services/integration executive, not a satellite lifer. That is deliberate and on-strategy: his entire thesis is to convert SES from a capacity wholesaler into a managed multi-orbit services + government business. Named 2025 Satellite Executive of the Year and a 2026 Wash100 winner — well-regarded in the industry.
• Track record: the defining act is executing the Intelsat acquisition — strategically logical (scale, spectrum, government book, synergies) but financially aggressive (3.9x leverage, dilution of the formerly-clean balance sheet, immediate net loss). The verdict on him is unwritten and hinges entirely on synergy delivery + deleverage. He inherited the O3b mPOWER mess; he didn't create it.
• Tenure & skin in the game: ~2 years in. Insider ownership is structurally minor — the controlling shareholder is the Luxembourg State (see governance below), not management. n/a — specific insider-ownership figures not sourced (insider-transactions.csv not on disk).
• Capital-allocation history: the big call was spend €3.1bn + take on Intelsat's debt to buy scale. Supporting moves: cut the dividend to fund deleverage (disciplined), cancelled 2 software-defined GEO satellites as a synergy move (disciplined), instituted a board-level CapEx taskforce (disciplined), and is fighting hard for the O3b mPOWER insurance recovery. The pattern post-deal is conservative and deleverage-focused — the right posture, but it confirms there is no organic-growth lever to pull in 2026 ("stable" guidance).
• Red flags (governance, not fraud): the dual-class structure with Luxembourg-State control is the standing governance issue. B-shares = 33.33% of votes but only 16.67% of economics. The State nominates 4 of 12 directors and can block/steer strategy for sovereign/industrial-policy reasons that may not maximise minority-shareholder value — e.g. SES leading the IRIS2 sovereign-EU constellation is strategically prestigious but is a politically-driven, capital-intensive, lower-return commitment a purely commercial operator might not make. Not a fraud flag — a principal-alignment flag.
• Archetype: professional manager / turnaround integrator, not founder-owner. Appropriate for this stage (integrate, deleverage, reposition) — but it means the upside is operational execution, not visionary value-creation.

Lens 10 · Forensic Red Flags

Forensic lens. Every figure ``; SES files IFRS as a foreign issuer.

• Reported vs. like-for-like — the biggest "watch" item. Reported revenue +33.9% while like-for-like is −1.6%, and reported EBITDA +19.1% while LFL is −12.1%. This is legitimate (Intelsat consolidated mid-year) and SES discloses both — but it is exactly the kind of optics where a headline reader sees "+34% growth" and misses that the underlying business shrank. Always anchor on LFL for SES.
• Adjusted vs. reported EBITDA gap. SES's "Adjusted EBITDA" strips out U.S. C-band repurposing income, M&A costs, specific business taxes, one-off regulatory charges — and SES swung to a €95m net loss on a reported basis once higher D&A, financing costs, and non-cash impairments from the acquisition flow through. The €1,196m Adjusted EBITDA is real cash-generative capacity, but the distance from "€1.2bn Adj EBITDA" to "−€95m net profit" is large and driven by genuine capital-intensity (D&A) + leverage (interest) — not to be hand-waved.
• The O3b mPOWER insurance claim (the sharpest forensic flag). SES booked / is pursuing a ~$472m insurance claim (70% of insured value) on the first four Boeing MEO satellites for the power-module defect. Insurers are resisting, arguing it's "a forecast, not an observed loss," and only ~$87m had been collected by Sep-2025 — over a year after filing. Watch how the unrecovered balance is carried (receivable vs. written down) and whether the satellites' reduced lifespan triggers future impairment.
• Hybrid notes (€1.525bn) — equity-credit risk. The three perpetual/long-dated hybrids get partial equity treatment from rating agencies; an adverse re-classification would raise reported leverage. Worth monitoring.
• Goodwill/intangibles. A $3.1bn acquisition of a declining-core business creates goodwill exposure. Search did not surface a specific FY25 goodwill-impairment figure (n/a), but "non-cash impairments linked to the acquisition" were cited in the net-loss bridge. This is the #1 thing to verify against the 20-F on the next hybrid-grounding pass — a future impairment is a real risk given the LFL EBITDA decline.
• Restricted cash: €401m of the cash balance is restricted (IRIS2 consortium) — so "€674m cash" overstates usable liquidity; the unrestricted figure is what matters for the 3.0x deleverage path.

Regulatory findings (required sub-section). Per regulatory/regulatory-findings.md (generated 2026-06-30): SES has no CIK in the local index → no SEC EDGAR (LR/AAER) enforcement search was run from disk; total_sec_findings: 0. Web check — "SES" satellite (FTC OR DOJ OR FCC OR settlement OR fine OR penalty) enforcement — surfaced no material enforcement action against SES S.A. the satellite operator (results were dominated by an unrelated "SES AI" battery company facing shareholder legal probes — not this company; do not conflate). SES's regulatory exposure is constructive, not adversarial: FCC C-band clearing payments (in its favour), ITU coordination, and EU IRIS2 concession award. The Intelsat deal itself cleared FCC and EU antitrust review (approved Jul-2025). Item 3 (Legal Proceedings) from a 10-K/20-F could not be quoted (filings not on disk — n/a — not on disk; verify on next ingest pass). Net: no material regulatory or legal findings against SES S.A. the satellite operator — verified via the regulatory-findings file (0 SEC, no CIK) + web search as of 2026-06-30; 20-F Item 3 not yet ingested.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

SES is currently loss-making at the net line, so a clean EPS path is unreliable; the management lever is EBITDA + deleverage + FCF, not EPS. I anchor on management's own LFL guidance and synergy plan and show the arithmetic. All ; inputs .

Anchors: FY25 LFL revenue ~€3.5bn, LFL Adj EBITDA ~€1.5bn (€1,529m), capex ~€700m, Adj FCF €229m (reported, ~5.5mo Intelsat), Adj Net Debt €6,029m, synergy run-rate target €370m (70% within 3 years), 2026 guided "stable."

Base case (FY26–FY28) — synergies offset core decline; the business stabilises but doesn't grow:

• Revenue: ~€3.5bn → ~€3.45bn → ~€3.5bn.
• Adjusted EBITDA: ~€1.50bn → ~€1.60bn → ~€1.70bn.
• Capex: ~€700m → ~€700m → ~€650m. Capex-heavy through 2027.
• Net leverage: 3.9x → ~3.3x → ~2.8x. This is the real "EPS" of the SES story.

Bull case: synergies land faster (full €370m by FY27) + a new C-band clearing payout materialises + government/IRIS2 revenue ramps ahead of plan → EBITDA toward ~€1.8bn by FY28, leverage under 3.0x by FY27, dividend raised, stock re-rates from 7.4x toward ~9x EV/EBITDA.

Bear case: Fixed-Data/Media decline accelerates (−10%+/yr as Starlink takes enterprise + maritime) and out-runs synergies; a goodwill impairment lands; insurance claim written off; EBITDA slips toward ~€1.3bn, leverage stays >3.5x, dividend stays suppressed, multiple compresses toward Eutelsat's ~6x or below.

Tracked forecast (Brier): the scoreable base call is "SES FY2026 Adjusted EBITDA ≥ €1.45bn (LFL, constant FX), p≈0.62, resolves 2026-12-31." Per the --watchlist rule, the forecast.ts create step is skipped in the loop — logged here in-dossier only; promote to a tracked forecast only on a committed /thesis pass.

Lens 12 · Bull vs Bear

Bull case. SES is the last independent, sovereign-aligned, multi-orbit Western capacity stack — and that scarcity is the asset. Post-Intelsat it has the spectrum, the GEO+MEO fleet, the government franchise growing double-digits (+17.3% LFL), and the IRIS2/EU-sovereignty mandate that Starlink structurally cannot fill (a US-controlled, single-vendor, single-orbit network is disqualified from sovereign-resilience roles). €370m of synergies + episodic C-band cash de-lever the balance sheet to <3.0x within ~18 months, the dividend gets raised, and a 7.4x EV/EBITDA "melting-ice-cube" multiple re-rates as the market realises the Networks/Government engine is durable and the media decline is a known, contained, cash-generative runoff. You're buying ~€1.5bn of EBITDA and a government growth call for an EV of €8.8bn while everyone is distracted by Starlink.

Bear case (the 2–3 things that can permanently impair). (1) Secular capacity deflation is faster than synergies. Data-capacity pricing fell ~−77% over five years (−26% CAGR) and video −16% — if SES's legacy ~€1.5bn book deflates 10%+/yr, no €370m synergy plan saves group EBITDA from sliding. (2) The balance sheet is the bet. At 3.9x leverage with a loss-making P&L, any EBITDA disappointment (or a goodwill impairment, or the insurance claim written off) pushes leverage the wrong way, keeps the dividend frozen, and risks a credit-rating downgrade that raises the ~4% cost of debt. (3) Starlink/LEO + direct-to-device keep expanding into SES's mobility/enterprise core; SES's LEO answer is partnerships (Lynk/Omnispace), i.e. it does not own the disruptive layer.

Pre-mortem (18 months out, thesis broke): It's end-2027. Synergies came in late and smaller than €370m; Fixed-Data and Media each fell double-digits as Starlink Enterprise + maritime ate the renewals; SES took a goodwill impairment on Intelsat; the O3b mPOWER insurance claim was settled for cents; leverage is still ~3.7x; the dividend was not raised; a rating agency moved SES to negative outlook. The stock is at Eutelsat's ~6x and made a fresh multi-year low. The autopsy line: "scale didn't fix a deflating core — it just added leverage to it."

Are multiples too high? 7.4x EV/EBITDA is not demanding for ~45% EBITDA margins and a government growth franchise — if EBITDA has stopped falling. It is too high if LFL EBITDA keeps declining double-digits, because then the "E" shrinks faster than the multiple looks cheap. The multiple is a leverage call on whether FY25 was the LFL EBITDA trough.

Contrarian view (what the market refuses to see): The consensus treats all GEO/MEO operators as one melting ice cube. The market is under-pricing the government/sovereignty bifurcation — Government + IRIS2 + assured-MEO is a different, growing, Starlink-immune business stapled to a declining media/data runoff. If SES can be valued sum-of-the-parts (a growing sovereign-connectivity franchise + a cash-generative media runoff) rather than as one shrinking blob, there's hidden value. But the same dual-class Luxembourg-State governance that guarantees the sovereign role also caps the multiple, because minority holders can't trust capital will be allocated to their return over industrial policy. The contrarian long and the structural discount share the same root cause.

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

Dismantling the bull case.

• What structurally breaks the model: the unit of sale (satellite capacity) is in structural price deflation (−26% CAGR in data) because LEO/HTS supply is exploding. SES's pivot to "managed services" doesn't escape this — it just moves up a stack that still rests on deflating capacity. A wholesaler in a commoditising input is the textbook value trap.
• Revenue concentration & shift risk: the growth is concentrated in US Government (~28%). That's a single buyer exposed to US budget cycles, procurement politics, and "buy American"/Starlink-favouritism risk (the US government is also Starlink's biggest booster). If a future US administration steers defence satcom dollars to SpaceX, SES's one growth engine stalls — and a European operator has limited recourse.
• Why the moat is weaker than bulls think: customers increasingly run dual networks (SES + Starlink) and use Starlink as the price anchor on every renewal. The "multi-orbit integrator" moat is partly SES re-selling/relaying other people's LEO (Lynk) because it doesn't own a competitive LEO of its own — a thin moat that erodes as Starlink/Kuiper add enterprise features.
• Most dangerous competitor bulls underestimate: not Starlink (priced in) — Amazon Kuiper, a second deep-pocketed LEO entrant with AWS-native enterprise/government distribution. Two subsidised mega-constellations is worse than one. And EchoStar's pivot (selling spectrum to SpaceX for $19.6bn + a SpaceX stake) signals even incumbents are choosing to monetise and exit rather than fight.
• Worst capital-allocation / governance: buying a declining business (Intelsat) with debt at 3.9x leverage to chase scale, while the controlling shareholder (Luxembourg State) can steer capital into low-return industrial-policy projects (IRIS2). The O3b mPOWER program — a Boeing satellite defect plus a year-long, mostly-unpaid $472m insurance fight — is a live example of capital and execution risk biting.
• Assumptions that must hold for today's price: (a) FY25 was the LFL EBITDA trough; (b) €370m synergies actually land; (c) no goodwill impairment; (d) US government spend keeps growing and keeps coming to SES; (e) leverage falls below 3.0x on schedule. Break any one and the equity (a thin ~€2.8bn slice under €6bn of net debt) is highly geared to the downside.
• Growth disappoints 20–30%: because the equity sits under €6bn net debt, the EV/EBITDA optics understate equity risk. A ~20% EBITDA shortfall (to ~€1.2bn) at a compressed ~6x EV/EBITDA → EV ~€7.2bn → equity ~€1.2bn after net debt → roughly −55% downside to the equity. The leverage cuts both ways and brutally so.
• Single scenario that permanently impairs: a credit-rating downgrade triggered by a goodwill impairment + EBITDA miss, which raises the cost of the heavy debt load, forces a dilutive equity raise or asset sale, and converts a deleveraging story into a balance-sheet-repair story. Plausibility: moderate — not the base case, but far from tail given 3.9x leverage on a declining core.

Lens 14 · Management Questions (ordered by information value)

1. Was FY2025 the trough for like-for-like Adjusted EBITDA, and what specifically inflects it positive in 2026 given you guided "stable"? (The whole thesis lives here.)
2. On the €370m synergy run-rate: how much is in the FY26 P&L, what's the quarterly cadence, and what's the realisation risk on the back half?
3. What is your hard timeline to net leverage <3.0x, and what's the contingency if EBITDA or C-band/insurance cash slips — would you raise equity or sell assets before cutting the deleverage target?
4. Have you taken — or do you foresee — any goodwill impairment on the Intelsat acquisition, and what EBITDA trajectory would trigger one under your IFRS testing?
5. What is the renewal pricing trend in Fixed Data and Maritime — quantify the price decline on contracts renewing in 2026, not just the volume.
6. On the O3b mPOWER $472m insurance claim: what have you actually collected, what's carried as a receivable vs. written down, and what's your realistic recovery estimate?
7. How do you defend the US Government franchise against a US administration that openly favours Starlink for defence satcom — what's your win-rate trend and pipeline concentration?
8. IRIS2: what return on capital do you underwrite on your ~€500m+ commitment, and how do you reconcile a sovereign-policy project with minority-shareholder value?
9. meoSphere (2030): total capital envelope, and how do you avoid a second O3b-mPOWER-style execution/anomaly risk with a new platform?
10. Where does direct-to-device revenue actually accrue to SES vs. to Lynk/Omnispace — are you the disrupted incumbent paying for relay, or a genuine beneficiary?
11. What is the realistic end-of-life / replacement capex on the legacy Intelsat GEO fleet over 2027–2030, and is it funded within the ~€700m envelope?
12. On the hybrids (€1.525bn): what's the refinancing plan at the call dates, and how exposed are you to a rating-agency reclassification raising reported leverage?
13. Given the dual-class State-controlled structure, what concrete governance protections exist so capital isn't allocated to industrial policy over returns?
14. What is your honest 5-year revenue mix target by 2030 (Government / Mobility / Media / Fixed Data) — i.e. how small does the declining book get?
15. At 7.4x EV/EBITDA and loss-making, why is buying back stock or paying down debt not a better use of capital than any new constellation investment beyond what's already committed?

---