Dead End: Iridium

The Call from the Bahamas

The origin story Motorola tells begins in 1985, when Karen Bertiger, on vacation in the Bahamas, could not make a phone call to close a real-estate deal and asked her husband why there was no phone that worked everywhere. Her husband was Bary Bertiger, a Motorola satellite engineer. In late 1987 Bertiger, with colleagues Raymond Leopold and Ken Peterson, worked out a design for global coverage from low Earth orbit, and Motorola filed the patents in 1988.

The design was ambitious in a specific way. Earlier satellite-phone concepts used a few large satellites parked in high geostationary orbit, which meant long signal delays and weak links to handheld devices. Iridium instead used many small satellites in low orbit, close enough for a handset to reach, moving fast enough that dozens were needed for constant coverage. The first plan called for 77 satellites, and the project was named for iridium, the element with atomic number 77. When the engineering was refined down to 66, the name stuck even though 66 is the atomic number of dysprosium.

A Routed Network in Orbit

The finished constellation was 66 active satellites (plus in-orbit spares) at about 781 kilometers, arranged in six orbital planes of eleven, near-polar so the paths crossed over the poles. What set Iridium apart was that the satellites talked to each other. Each spacecraft held Ka-band links to its neighbors, so a call could be routed satellite to satellite across the sky and dropped down to a ground station far from where it started. It was a packet-routed mesh network in space, one of the more complex distributed systems built in the 1990s, and it worked. Motorola built it under a fixed-price contract between 1993 and 1998 at a cost near $5 billion.

Service opened on 1 November 1998. The ceremonial first call went from US Vice President Al Gore to Gilbert Grosvenor, a great-grandson of Alexander Graham Bell.

Dead End

The business was broken before the first satellite launched, for a reason no engineering could fix: while Iridium spent the 1990s in development, ordinary terrestrial cellular networks spread across the populated world far faster and cheaper than the plan had assumed. Iridium was selling a solution to a coverage problem that was disappearing everywhere its likely customers actually lived and worked.

What it offered instead was expensive and awkward. A handset weighed about a pound and cost roughly $3,000. Calls ran several dollars a minute. The phone needed a clear line of sight to the sky, so it did not work indoors, in a car, or in a city canyon, the places people make most calls. Iridium had modeled hundreds of thousands of subscribers; it signed up a tiny fraction of that. On 13 August 1999, nine months after the ceremonial call, Iridium filed for Chapter 11 with more than $4 billion in debt, among the largest bankruptcies of its era. Motorola, having failed to create the mass market it imagined, announced in 2000 that it would deorbit the constellation, a controlled reentry that would have dropped all 66 satellites into the atmosphere to burn.

The Afterlife

The satellites were never deorbited. In December 2000 the US government stepped in with a $72 million contract for two years of unlimited access, and in March 2001 a group led by aviation executive Dan Colussy bought the entire system out of bankruptcy for about $25 million, a purchase that erased over $4 billion of debt and founded Iridium Satellite LLC. Service resumed in 2001, aimed now at the customers who genuinely had no alternative: soldiers, sailors, pilots, oil crews, scientists in Antarctica, and journalists in places without towers. The Department of Defense, which needed communications that did not depend on any host nation’s ground infrastructure, became an anchor customer and still runs its own Iridium gateway.

The company went public through a 2009 merger (NASDAQ: IRDM) and then did the thing that had killed it in reverse: it replaced the constellation while it was still flying. Iridium NEXT, 75 satellites launched on SpaceX Falcon 9 rockets between 2017 and 2019 at a cost around $3 billion, swapped out the entire first generation one orbital plane at a time. The new satellites carried hosted payloads that turned the network into infrastructure other industries depend on: Aireon tracks the ADS-B transponders of aircraft worldwide, giving air-traffic controllers surveillance over oceans for the first time, and a maritime payload does the same for ships. Iridium passed a million subscribers in March 2018 and runs profitably, and its network delivers the satellite SOS messaging in handheld devices like Garmin’s inReach line.

The lesson is not that Iridium’s engineers were wrong. Their machine did exactly what they designed it to do, and it is still doing it. They were wrong about the market and the timing, building a $5 billion consumer product for a coverage gap that cheap cell towers were closing underneath them. The cheap-launch economics that later made Starlink’s far larger constellation viable were exactly what Iridium lacked in the 1990s, when every satellite rode an expensive rocket. Motorola built the right machine on a business plan that the spread of cheap cell towers had already overtaken. The full corporate context sits in Motorola.

📚 Sources

• Iridium satellite constellation — Wikipedia — the 1987 concept, the 77/66 naming, constellation design and inter-satellite links, the $5 billion cost, and Iridium NEXT
• Iridium Communications — Wikipedia — the November 1998 first call, the August 1999 bankruptcy, the deorbit plan, the $72 million DoD contract, the $25 million rescue, and the 2009 public listing
• Finkelstein, Sydney and Sanford, Shade H.: “Learning from Corporate Mistakes: The Rise and Fall of Iridium” — Organizational Dynamics 29:2 (2000) — the business-case autopsy of the market and timing failure