From time to time I have revisited this concept mostly with reference to other I-in-the-sky projects.
A similar project was announced a while back using drones. I do not recall the details, but I have not heard much about it since.
What has caught on, however, are LEOs (now called Non-geostationary Orbit (NGSO) Systems). I was in the business when Motorola launched its Iridium constellation. In fact, I even bought stock in it. Then, eventually, I ended up using that stock certificate to light my fireplace.
The move to capture a piece of the internet pie has driven some companies to try some interesting concepts. Drones, balloons, NGSO systems and more. I will address the NGSO market a bit further on. But, first I want to dissect the Loony project’s failure.
Loon has had some success. It was able to put up a commercial system in Africa last year. As well, they have been successful in bringing temporary connectivity during disasters – Peru is one example. Another being in Puerto Rico after Hurricane Maria.
Why did Project Loon not succeed? For the same major reason that will plague NGSO systems going forward – market size and RoI potential.
Alphabet just could not come up with a model that was able to generate profits. They had decent commitments, but it was still challenging. Ninety-three percent of the world now has terrestrial Internet access of some sort. The other seven percent of places are mostly populated by people who either cannot afford the 4G phones that Loon requires or are simply not interested in getting access. There was just not a large enough market.
There was also the issue of longevity – five months was a balloon’s lifespan. Then there were regulatory issues around the globe and keeping them out of harm’s way as well, although that was minor. There was also the issue of sabotage. The biggest challenge, however, was from NGSO systems. Particularly the flamboyant Elon Musk and SpaceX.
But some of the LEO organizations have new tricks up their sleeves. For example, a newcomer, Facebook-owned PointView Tech, has a multi-million-dollar experimental satellite named Athena. It claims the E-band, in which it operates, will be capable of delivering data 10 times faster than SpaceX’s Starlink and other constellations.
The NGSO playing field is becoming crowded. In addition to the well-worn Iridium, other players include Telesat, Kepler Communications, LeoSat, OneWeb, O3b-SES, ViaSat, and, of course, the well-funded Starlink. While each may have a bit of a different scheme, they all want to be an OISP (orbital internet service provider) (I just made that up).
Now, back to the challenge of profitability. Recently, Joe Madden founder of Mobile Experts and one of the analysts I have a lot of confidence in, peeled back some of the challenges facing the NGSO industry as an ISP. He is also a speaker at our upcoming February 11th AGL Virtual Summit.
Mostly he discusses the economic models of NGSO systems versus terrestrial providers. His analysis makes a lot of sense, especially in light of the market becoming crowded. It will be difficult for each provider to carve out a niche to distinguish themselves from the pack. But in the end, the bottom line is all about profit.
Madden makes some salient points. I will recap a few here.
One of his points is that, back in the day, when Iridium launched, mobile communications minutes cost about $1.00. That was the data Motorola used to calculate the value proposition of the Iridium. However, as he points out, by the time the constellation launched, the cost had come down to about ten-cents per minute. At that point, they were already behind the eight-ball.
Motorola was looking ahead to data. However, in the meantime, the data model took too long to materialize and the project went bust. They had the right idea and eventually, 4G enabled the new data market. That is the model today’s NGSO systems players are counting on.
Terrestrial provider data costs are measured per gigabyte, nowadays. On a global scale, a gig of data ranges from less than $0.10 in India to over $50 per gig in Saint Helena (in the South Atlantic).
In North America, a gig of data runs between $10 and $30. Most of Europe and South America is between $1 and $10. Asia, Africa and the Russian Federation are mostly $10 and under. So where can the NGSO providers make money?
Around the world, there are places such as Saint Helena where the cost of data is substantial. There are not many but there are some. Those are potential markets.
As well, there are other situations, trains, planes and automobiles, and users where terrestrial coverage is sparse (ever driven I-70 across Utah?). There are specialty markets as well (government, infrastructure, agriculture, and variants of others).
It is safe to say that the NGSO systems cannot be competitive in well covered terrestrial markets. As the cost of data ramps up in expensive to cover areas (rural U.S., Africa, remote areas) the cost equations begin to equalize, even run in the black for NGSO systems. But there are not enough of them to spread across the players.
Madden’s data came up with a $5 – $8 per gig cost from these satellites. While currently there is a fair amount of opportunity in the over $8 gig markets, the cost is not always the driver. Today, satellites have unacceptable trip times for critical applications. The best numbers for latency are around 25 milliseconds (ms). As well, the bandwidth is narrower, thusly so is the speed of the data (Starlink is expected to deliver an average of 40 Mbps for the download with a latency of just under 40 ms).
However, there are many applications where latency is not necessarily a problem now, and perhaps even into the future. But for many real-time applications, such metrics are not acceptable.
This may work for specialty markets, as I pointed out earlier, because of their nature, or it is the only option. However, that is changing rapidly. With 5G on the horizon, global wireless availability will ramp up – and rapidly once 5G “gets real” as Verizon alluded to in one of its recent ads. That will put even more price pressure on the NGSO segment.
However, time and technology wait for no one. So, it is reasonable to expect that NGSO technology will improve. Over-the-air type upgrades will be the norm. And configuration issues and multiple configurations (both land and space integration) are also likely to take place. The terrestrial providers may see value in NGSO systems to expand their footprint.
Madden makes an interesting prediction. He believes that most, if not all, of these providers will go under because the terrestrial market will only get cheaper and coverage will only increase. I tend to agree. Once the terrestrial markets expand, especially with 5G, the specialty markets will see a contracting revenue base. It will become more difficult to cover the cost of the NGSO systems, even if they manage to optimize their costs.
The future of NGSO systems is certainly going to be interesting. How, and if, they will manage to run in the black will be what to watch. There is also likely to be consolidation at some point down the road, perhaps from bankruptcy, perhaps to prevent bankruptcy, perhaps simply because it will be beneficial for the parties involved. Then there is always the possibility that NGSO systems will collaborate with terrestrial providers.
In the end, look for satellite internet to simply be a specialty segment of the global wireless umbrella. However, I doubt I will be adding Iridium, or any other NGSO company stocks, to my portfolio in the future.