The satellite communications industry has spent the better part of a decade explaining, with tremendous confidence and very attractive slide decks, that it is in the process of reinventing how humanity connects. This is a compelling story. It is also, increasingly, a story that regulators, enterprise buyers, and a few impatient investors are starting to interrupt with a fairly basic question: yes, but is any of it working yet? The week of March 6 through March 13 was not kind to the parts of the industry still relying on the answer being self-evident. Concrete progress, it turns out, is a higher bar than the sector’s recent conference schedule implied.
LEO: The part where someone finally buys some land
On March 10, Telesat announced acquisitions and leases to support landing stations for the Lightspeed network, comprising a purchase near Estevan, Saskatchewan, a lease in Shaunavon, Saskatchewan, and a purchase near Papineauville, Quebec. The intent is to connect the future constellation into major terrestrial fiber infrastructure, with routing diversity built across geographically separated sites, in support of what the company describes as sovereign digital services for northern and Arctic Canada. Telesat does not expect to launch its pathfinder satellites until December 2026, and global commercial service involving around 24 landing stations remains further out still. So the announcement is, in a narrow sense, about purchasing fields.
What it is actually about is the growing awkwardness of selling a network that exists primarily in investor materials. Governments and enterprise customers have developed a mild allergy to connectivity services that are perpetually one milestone away from being connectable, a condition the industry has somehow normalized as a business model. By making ground infrastructure visible now, parcels with fiber access logic and a clear geographic rationale, Telesat is attempting to demonstrate that Lightspeed is a system being built rather than a concept being maintained at room temperature. Whether that lands depends on whether the December pathfinder holds. But as a positioning move in a market crowded with orbital ambition and thin operational evidence, it is smarter than another rendering.
The LEO story also picked up an uncomfortable subplot from Washington. FCC Chairman Brendan Carr publicly raised concerns on March 12 about Amazon’s pace of Kuiper constellation deployment, specifically around milestone compliance attached to the company’s license. Amazon has managed to operate in relative quiet compared with Starlink, which is a reasonable strategy until the regulator starts asking questions on the record. Carr’s remarks did not constitute a formal enforcement action, but they placed Kuiper in a position it had not previously occupied: under named scrutiny, with a public clock. For a company that has so far preferred to develop its constellation at its own disclosed-but-not-debugged pace, that is a materially different environment, and not one Amazon appears to have been eagerly anticipating.
D2D: The IoT angle nobody is talking about enough
The D2D conversation this week mostly belonged to Hubble Network, which on March 9 at Embedded World demonstrated a two-way connection between a standard, unmodified Bluetooth Low Energy chip and a LEO satellite. No custom hardware. No modified firmware. The kind of BLE chip already embedded in billions of low-power devices across every industry that has ever needed to track something remotely, made satellite-addressable without touching it. The company’s seven-satellite constellation is operational, and it opened a commercial developer portal the same week, apparently confident that the industry had matured enough to find this useful rather than just impressive.
What Hubble Network is addressing is the device population that doesn’t carry a SIM and was never going to: the installed base of ordinary IoT hardware that has always been limited to its local radio range. That population is large enough to matter. Standard BLE chips are in agricultural sensors, cold-chain trackers, asset tags, environmental monitors, and a long list of other devices deployed in locations where cellular coverage is either absent or economically unjustifiable to build. The pitch to developers is straightforward: use the hardware you already have and get satellite connectivity you would otherwise need a specialist terminal to achieve. Whether that holds up at scale depends on the link budget, which Hubble has not yet had to prove under commercial load.
The question of how it scales commercially is the interesting one the developer portal will start to answer. It will also start to reveal whether Hubble’s seven-satellite constellation is sufficient to meet the revisit rates that actual IoT use cases require, or whether the hardware story has outrun the network behind it, a trap the satellite industry has walked into before with some enthusiasm.
GEO: Deprecated in theory, collecting defense contracts in practice
On March 9, SpaceX launched EchoStar-25 for Dish Network into geostationary orbit on a Falcon 9. The launch was uneventful, which is the preferred outcome. Dish gets additional broadcast capacity during a corporate period best described as complicated. The wider industry gets a data point about GEO’s installed-base logic remaining commercially active while everyone argues about whether it should be.
Also during the period, Lockheed Martin and Mitsubishi Electric announced an agreement to deliver an anti-jamming communications payload for Japan’s next-generation GEO defense communications satellite. Defense customers buying resilient, survivable GEO capacity in the Indo-Pacific is not surprising, but the contract is a useful corrective to the idea that GEO is quietly being deprecated by people who have read too many LEO pitch decks. High-consequence users tend to buy what works reliably under adversarial conditions, not what won the most recent conference panel about orbital architecture. GEO continues to pass that test in ways LEO constellations are still in the process of demonstrating.
The standards dimension of GEO’s week ran slightly before the period opens. Gilat’s end-to-end 5G NTN demonstration over GEO was announced February 26 and remained commercially active in conversations throughout this week. The demonstration mattered because it pushed back against a fashionable assumption: that standards-aligned satellite networking is naturally a LEO story and GEO should be grateful for any role it gets to play. Optimizing 5G NTN for GEO’s actual latency characteristics, rather than treating them as a permanent disqualifier, is a more interesting engineering choice than orbit tribalism usually allows. Relevance maintained through adaptation is still relevance, even if it generates fewer keynote invitations.
Defense and spectrum: the regulatory week that got underreported
Lockheed Martin announced on March 9 a commitment of more than £100 million into the UK space sector. The detail worth sitting with is not the size of the number but what it is responding to. The UK has spent several years building a policy environment that is regulatory and industrial in equal measure, with a defense orientation baked in, explicitly designed to make the country a credible anchor for space investment rather than a spectator of decisions made elsewhere. Lockheed’s commitment suggests that architecture is producing results. For satcom operators watching where large defense contractors place long-term bets, the direction of travel is legible: toward jurisdictions that have done the regulatory and industrial groundwork, rather than toward whichever orbital market is currently generating the most enthusiasm at trade shows.
The FCC’s activity between March 10 and 12 carried different stakes and received considerably less coverage than it warranted. Space Bureau Chief Jay Schwarz described several proceedings aimed at opening satellite spectrum for non-communications missions, with a vote scheduled for March 26. More consequentially, WRC-27 preparation surfaced publicly at a Congressional event on March 12. WRC-27, the World Radiocommunication Conference scheduled for 2027, is where international spectrum allocations for the next generation of satellite services will be negotiated and locked in. What gets agreed there will determine which constellation architectures are viable at scale and on what terms satellite and terrestrial systems are permitted to coexist globally, including how much room remains for new entrants in the bands that actually matter. The conference is two years away, which is precisely why the pre-negotiation positioning happening now matters. Operators who arrive at WRC-27 without having shaped that landscape tend to find it has been shaped without them.
VSAT and service continuity: inheriting the 2 a.m. call
The satcom industry runs on announcements about the future and tends to treat the present tense as a waiting room. SKYTRAC’s March 11 news was resolutely present tense: the company announced it had completed the transition of Honeywell’s Sky Connect customer base, assuming full support and maintenance responsibility for an installed fleet of aviation terminals. No satellites involved. No orbital mechanics. Just an agreement to be the company that answers when something stops working, which in aviation satcom is the agreement that actually matters.
The company that handles service continuity controls the upgrade path. Operators do not primarily evaluate their connectivity provider on architectural elegance when the link drops over the Atlantic. They evaluate whoever picks up the phone and fixes it. SKYTRAC now occupies that position across a meaningful slice of the installed aviation satcom market, acquired without a single orbital asset, through the old-fashioned mechanism of agreeing to be responsible for something. The industry has no keynote slot for that. The customers filing the next RFP do.
The VSAT story more broadly is that operational discipline does not become worthless because the orbital hardware above it gets more interesting. Newer architectures import the same requirements they were supposedly replacing. What changes is the branding on the equipment.
Infrastructure: the unglamorous layer that scales everything
SSC Space announced on March 10 a new optical ground station in Santiago, Chile. Optical ground infrastructure does not produce the kind of announcement that gets a countdown clock or a webcast. It produces a press release describing a building in South America, which most sector coverage filed under things that will matter later and moved on from.
That instinct was wrong. Higher-capacity constellations and more demanding applications need more capable paths between orbit and Earth. Optical links offer bandwidth and security characteristics that conventional RF ground networks struggle to match at the performance levels operators are now promising with considerable public confidence. SSC Space’s Santiago node extends Southern Hemisphere coverage in a network that is becoming a strategic layer in its own right, something operators and governments will compete for access to, not just a technical amenity that appears in slide appendices. The firms that end up mattering in the next phase of satellite communications will not all be counting spacecraft. Some will be the ones that control where data can reliably leave orbit, and optical ground infrastructure, once built and positioned, is not easy to replicate quickly. That makes it a more durable competitive asset than the satellite valuations it currently supports would suggest.
What the week actually said
What this week surfaced is that the industry’s evaluative standard is shifting in a direction the industry did not entirely choose. Constellation aesthetics and theoretical capability are no longer sufficient as answers. Buyers and regulators are now asking who has the authorization, what the deployment timeline actually is, what happens when service fails, and whether the spectrum plan is realistic. These are duller questions than the ones the sector has been fielding at panels for the past several years, and they are considerably harder to answer with a well-designed logo.
They are, however, the questions that infrastructure markets ask when they are deciding what to trust. The satellite communications industry is discovering, with varying degrees of grace, that it has graduated into that category, and that the graduation comes with coursework nobody signed up for.
