Iridium

Iridium is the oldest operator in this directory, the only one that has completed a full constellation replacement, and — pointedly — the only one that reliably makes money.

It does not sell broadband and does not try to. It sells something the broadband constellations cannot: a global, weather-resilient, low-bandwidth link that works at the poles, in the middle of the Pacific, and in conditions where a Ku-band dish is useless. Its customers are ships, aircraft, pipelines, remote sensors, expedition teams and the United States Department of Defense.

The Cautionary Tale Everyone Forgets

Iridium is the original satellite-constellation bubble, and its history should be required reading for anyone underwriting the current one.

Conceived at Motorola in the late 1980s, it cost around $5 billion to build and reached operational service in November 1998. It filed for bankruptcy nine months later. The technology worked. The market did not exist — the handsets were enormous, the calls were expensive, and terrestrial mobile coverage expanded far faster than anyone had modelled.

In 2001, investors bought the entire constellation, the ground infrastructure and the spectrum for roughly $25 million — around half a percent of what it cost to build. The company that emerged went public in 2009 and has been profitable and cash-generative for years.

The lesson is not that satellite constellations fail. It is that the first owner of the infrastructure frequently does not capture the value. Capacity gets built with someone else’s money and the business gets made by whoever buys it at the bottom. It is worth holding that thought while reading any LEO valuation model.

The Constellation

Metric	Position	Source type
Operational satellites	66, plus spares (75 launched)	Company statement
Orbit	780 km, 6 polar planes	Regulatory filing
Inter-satellite links	Full mesh cross-links — since 1998	Technical specification
Replacement programme	Iridium NEXT, completed January 2019, ~$3bn	Company filings
Launch provider	SpaceX — all 75 NEXT satellites, Falcon 9	Launch record

Two things deserve attention. First, Iridium had a fully cross-linked mesh network in 1998 — the capability the industry now treats as a Starlink innovation and which OneWeb’s first generation still lacks. Traffic routes satellite-to-satellite and can reach the ground at a small number of gateways, which is precisely why the network works over oceans and poles.

Second, Iridium NEXT is done. The replacement fleet was completed in January 2019 for around $3 billion, and the satellites are expected to serve well into the 2030s. Alone among the operators here, Iridium has its enormous capital expenditure behind it rather than in front of it — which is the entire reason it converts revenue into free cash flow while the others raise money.

Business Model

• IoT — the growth engine and the majority of billable subscribers. Trackers, sensors, fleet and asset monitoring: low bandwidth, low churn, high margin.
• Government — a long-running Enhanced Mobile Satellite Services contract with the US Department of Defense, providing dedicated, gateway-isolated capacity. Contracted, sticky, and very hard for a competitor to dislodge.
• Maritime and aviation — including GMDSS safety-of-life services, where Iridium is one of only two authorised providers. Regulatory approval of this kind is a moat that money alone cannot buy.
• Iridium Certus — mid-band broadband, competing with VSAT at the lower end.

Recent annual revenue has been in the region of $800–850 million, with billable subscribers in the low millions and IoT the dominant and growing share. Verify against the current 10-K.

The Starlink Problem

The bear case is simple, and it is not stupid: direct-to-device services from Starlink and AST threaten to make a dedicated satellite handset unnecessary, and Starlink’s maritime and aviation products already undercut Certus on cost per bit by a wide margin.

The bull case is that Iridium sells something structurally different. L-band works through rain, foliage and heavy sea state, where Ku-band degrades. Safety-of-life certification takes years and cannot be bought. Government customers value a network with a sovereign, isolated gateway architecture and no consumer traffic on it. And an IoT sensor on a pipeline needs a few kilobytes a day with total reliability — not 200 Mbps.

Iridium’s answer is Iridium NTN Direct, a standards-based NB-IoT direct-to-device service built on 3GPP non-terrestrial network standards, which lets ordinary chipsets reach the constellation without proprietary hardware. It is a defensive move into exactly the territory Starlink is attacking, and whether it lands is the central question for this company.