As the satellite IoT sector has been evolving at a rapid pace in the past months, I thought it was time to give an updated overview of the sector.
Let’s start with the French rising star, Kineis. After a delay of about a year, launches will happen this year until 2025 to establish a constellation of 25 satellites. This will allow Kineis to start delivering its connectivity service globally and get a leadership position being the first fully operational IoT satellite constellation. The key issue it will be facing – like all other contenders – is how to maximize revenues over the lifetime of the infrastructure. The traditional connectivity model (a few €/month with indirect sales and a slow ramp-up) may not be the best way to go especially considering the global competition.
Historic players of the first wave of satellite IoT are not in a good position anymore. Dutch Hiber has changed its business model and will not launch satellites but focus on downstream markets as a solution provider for O&G. Same for Australian Fleet Space that pivoted to EO with a “mineral detection” service provided by its satellites (forget abour Lora!). Fellow Australian Myriota is still in the race with operational services in many countries (Americas, Europe…). While the number of satellites in orbit is unclear (probably less than 10), Myriota has the traditional IoT market approach of a low-cost operator proposing its services to an ecosystem of solution providers. Swiss based Astrocast benefits from a good spectrum in the L-band and support from Airbus and Yahsat. With 18 satellites in orbit it can already provide good quality of service. Latest financial update (feb 24) present revenues below 0.2M CHF for the first half year of 2023 and a loss of 10M CHF. The company probably needs to raise additional funds with a cash burn rate estimated over 1M CHF/month. The main issue is on the revenue side: at 1 CHF/month/device, this means that there are between 20000 and 40000 devices generating revenues (if revenues are only from subscriptions). Should we take into account other revenues like hardware, training, etc… this number can be divided by 2. Clearly, the ramp-up is way too slow and is does not demonstrate a high adoption rate….
Lora (a terrestrial IoT communication protocol supported by Semtech) has been adapted for Satcoms and is used by several companies. The benefit of Lora is the access to a large ecosystem of developpers, operators and marketeers Lacuna Space was the first to launch its satellites a few years back. Apparently the company is provisioning new satellites from Spire and its is unclear at this stage how many operational satellites they have. The company is focusing on technology and sells via partnerships with other operators (Omnispace, Wyld networks…), A few new LEO players recently emerged. Italian Apogeo just launched 9 pico satellites with 9 more in the pipe. With a simple price structure, it hopes to grow quickly. Its main advantage: very reduced costs thanks to pico satellites. Their model is similar to US based Swarm which was bought by Starlink (probably the best future for Apogeo would be to be integrated in another large operator). Fossa systems is a more recent newcomer based in Spain. An engineering company, it proposes also Satellite IoT services,
But Lora is not the only standard as 5G (especially NB-IoT) can also be used for satellite IoT, with a certain complexity at the satellite level. The benefit here is the access to a huge ecosystem of cellular operators and industrials, No company has reached its market maturity yet. In Europe, Spanish Sateliot and Luxemburgish (?) OQ are the main companies, In the US, Lynk also proposes 5G base space IoT. These companies need to raise funds to launch their full constellations, which is probably a few years away. All have a similar gotomarket: partner with existing cell operators to provide extended connectivity outside of cell coverage to exting IoT devices. While NB-IoT has been slow to take off, it is now part of 5G and should surpass Lora or other protocols. In the end, probably a good technology choice and a simple business model that however does not allow to control distribution.
Legacy satcom operators are also trying space IoT. While French Eutelsat seems now to focus on OneWEB while they announced IoT plans in the past, others are more proactive. US Echostar has opened a European service in S band using Lora based on a GEO satellite. They announced US coverage (from GEO) and future global coverage with a LEO constellation, The gotomarket is the usual connectivity approach which is very slow to take off. While the service is working flawlessly (I have seen it in action) and almost in real-time (which is unique), the key question remains how to sell it quickly. British Inmarsat is also planning an IoT service using NB-IoT over its GEO satellites in L-band. This service is available from US-based Skylo using a new Mediatek chipset. Other LEO players like Iridium and Globalstar are considering the opportunity however there is more business to make in higher data rates as mature IoT/M2M markets are always asking for more data and more real-time for a higher price point.
And I forgot to mention e-space which plans to provide IoT connectivity as well with a massive constellation. However, there is not enough information available at this stage to have any idea of the services that will be provided.
I’d like to finish with the usual elephant in the room, Starlink, that bought Swarm a few years ago (picosats to provide satellite IoT in VHF) and announced an IoT service in 2025. Given its track record in space, we can probably expect a game changer in the field: Starlink revolutionized broadband satcoms (a real disruption: it is easy and cheap), is disrupting Direct-to-Device (they demoed satellite texting with plain vanilla iphones) and will probably completely change the satellite IoT ecosystem. How? Wait and see!
In the end, while I am convinced of the market opportunity for satellite IoT, history, both in the IoT world (slow take off) and the satellite world (only bankrupt companies succeeded) has shown that the major difficulty is in marketing and gotomarket. The operator model is too slow, especially when costs are high. So new models are needed and companies with low Capex (picosats or existing satellites) have definitively an advantage.