MDA Space Buys Its Raise With Two Points of Margin
A CAD 498M quarter, a bigger backlog question, and a CEO who stopped dismissing the launch worry
MDA.TO · Earnings Call · 2026-11-05
The beat, the raise, and the quiet cost
MDA Space posted Q3 2026 revenue of CAD 498 million, up 22% year-over-year, with adjusted EBITDA of CAD 96 million. Mike Greenley opened with the kind of line CEOs only get to use once: “When we went public in 2021, five and a half years from now, those numbers were our numbers for the whole year. Now we just did the same amount in the third quarter.” — Mike Greenley, Chief Executive Officer · 2026-11-05 Year-to-date revenue reached CAD 1.46 billion, up 29%, and guidance went up again — revenue to CAD 1.9–1.95 billion from CAD 1.8–1.9 billion, adjusted EBITDA to CAD 355–375 million from CAD 330–370 million. Look at the shape of the raise, though. Revenue growth at the midpoint moves to ~18%; EBITDA growth to ~13%. The company is guiding into margin compression, and Q3 already showed it: gross margin of 25.3% versus 26.4%, adjusted EBITDA margin of 19.3% versus 20.2%. Guillaume Lavoie was explicit that this was “a decrease compared to 26.4% in Q3 of 2025, driven by a change in program mix” — Guillaume Lavoie, Chief Financial Officer · 2026-11-05 — the phrase change in program mix is doing a lot of work in this story. Asked on the same call about the implied step-down to ~18% in Q4, the CFO was candid: “we are investing in scaling up our business. That is really, really important to us. I mean, we are not necessarily focused that much on the next 12 months.” — Guillaume Lavoie, Chief Financial Officer · 2026-11-05 That posture isn't new. Six months ago, on the March 2026 call, Lavoie said of defense-heavy mix: “at the end of the day, we don't take margin to the bank, right?” — Guillaume Lavoie, Chief Financial Officer (CFO) · 2026-03-04 Investors are now watching that philosophy compound across a second guidance cycle. Growth is real; the question is what the incremental dollar of CAD 498 million-quarter revenue actually earns.“Prime contractor” is back at the top of the list
The most striking keyword signal this quarter is prime contractor ranking first after being a notable decliner earlier in the year. The company is deliberately reframing itself from merchant supplier of subsystems to owner of missions — Globalstar's eight replenishment satellites were the first commercial LEO communications constellation MDA delivered as prime, with nine more to follow later this year, and Blue Canyon Technologies is explicitly the vehicle for an "MDA Space USA" that can bid as a U.S. prime. What's refreshing is the honesty about IRIS². Greenley called the Thales Alenia Space authorization-to-proceed for entry point exactly that — “the authorization to proceed for us to provide digital payload elements for the IRIS² constellation, I think that's a really strategic entry point into a large European communications constellation” — Mike Greenley, Chief Executive Officer · 2026-11-05 — and then, under pressure from National Bank's Doug Taylor, explained plainly that European governments will want European content first, and that MDA is "on the team" with ready-to-go digital payloads while the local industrial base catches up. That is a supplier position dressed as a prime position, and management said so out loud.The customer management stopped talking about
The most informative omission: Globalstar's parent-to-be is never mentioned. In May 2026, Greenley spent a full answer on Amazon's agreement to buy Globalstar — “it's early days, obviously, with the Amazon announcement to purchase Globalstar” — Mike Greenley, Chief Executive Officer · 2026-05-07 — walking through why conversations had to wait until the deal closed. This quarter, on a call that mentions Globalstar once, as nine satellites still to be delivered, Amazon disappears entirely. Meanwhile the Telesat Lightspeed backlog gets a hard number — roughly CAD 1.4 billion, per the CFO — while the Globalstar next-generation backlog never does. When a company quantifies one giant contract and not the other, that is the story. The Pathfinder satellites question was the analysts' proxy for the same anxiety: Telesat publicly pushed its launch timing and referenced software integration. The answer was that nothing in the financial forecast changed, and that milestone payments are deliberately sliced into small steps rather than lumpy completion bonuses — a structure that protects quarter-to-quarter cash but also means no single event rescues a weak cash year.The launch answer changed register
Compare two questions about SpaceX and launch access, ninety days apart. In August, Greenley was categorical: “We have not seen that yet, no. No. Like I've not been in a conversation where someone said we're adjusting our time lines, blah, blah, blah based on launch. That is not occurring, no.” — Mike Greenley, Chief Executive Officer (CEO) · 2026-08-07 In November, asked whether he loses sleep over Falcon 9 being sunset: “I'm not losing sleep yet, but it's a legitimate watch point, as you know.” — Mike Greenley, Chief Executive Officer · 2026-11-05 Same management, same question, different register. That is what a genuine new risk sounds like when it first enters a management team's language. The wider tape is voting the other way. Over 90 days the space-and-defense complex has bled hard — the Space Development Agency basket down meaningfully, the U.S. Space Force basket alongside it, and flight test program names down more than a third. MDA raising guidance into that de-rating is a real divergence, and it cuts both ways: either the market is mispricing a profitable, backlogged contractor, or it is telling you something about launch cadence, budget timing and the cost of sovereign content that management is only now starting to acknowledge.The mix shift that actually matters — and the cash it costs
The structural story is buried under the quarterly noise. Closing CLS would push recurring revenue from a minority of sales to over 30% of annual revenue, per management, and CLS plus BCT together would have added almost CAD 700 million of 2026 revenue on a pro forma basis. GeoIntelligence grew 25% in the quarter, and the MDA CHORUS launch — three satellites, C-band wide-swath plus X-band sub-meter, anchored by a 19-year-old RADARSAT-2 — is the first time MDA sells a service rather than a spacecraft. That is the difference between a project manufacturer and a data business, and it is why the order momentum of roughly CAD 1.6 billion in gross bookings through nine months matters more than the just-under-CAD-4-billion backlog, which has averaged flat for ten quarters. The bill comes due in cash. Free cash flow was negative CAD 116 million year-to-date against positive CAD 185 million a year earlier, with operating cash flow of CAD 74 million versus CAD 353 million, and the CFO flagged "a bit of consumption of cash" in Q4. Net cash of CAD 1.3 billion and total liquidity of CAD 2.9 billion are earmarked for the two acquisitions, not for shareholders. There is also a genuine new thread: drone work through 49North, with the RCN ISTAR program and the General Atomics RPAS team positioning MDA in Canadian autonomous systems for the first time in a decade — and analysts asked about it unprompted, which is usually the tell that a narrative is forming.The company is executing, and the revenue is not in doubt. What is in doubt is the price of that growth: two points of margin this year, more in 2027, in exchange for a services mix that won't show up in the numbers until CHORUS is commissioned and CLS is consolidated. That trade is defensible. It is also, for the third straight guidance cycle, the same trade.It is always great to have a beat and raise moment, and we will take them whenever we can get them. We will stay focused getting closed and be back in the new year to talk about how that ended.