Telcos take steps to monetize B2B satellite services

The spate of deals between satellite operators and telcos this year have largely focused on partnerships to deliver direct-to-device consumer satellite connectivity. But behind the headline-making stories, several telcos are looking at how to profitably build satellite connectivity into enterprise service offerings.

Today, satellite connectivity is firmly positioned as complementary to terrestrial mobile networks, rather than a replacement. One of the reasons for this is a simple matter of physics.

“Satellite connectivity requires plain sky to work properly and it is still unsure how it will work without a proper sky sight,” says Tom Sjöberg, Director, Business Change Management, at Finnish operator Elisa

There is also the matter of cost. Telcos’ competitiveness with services relies on them providing the best mix of ease-of use, coverage and cost-effectiveness. Currently, “terrestrial IoT connectivity – such as 2G, 4G, and [narrowband] IoT – outplays satellite connectivity in terms of data transfer price,” says Sjöberg.

This leaves satellite to fill in areas where there is no terrestrial coverage, such as oceans or rural areas, translating into limited revenue opportunities for communications service providers (CSPs), notably in the IoT services space.

How lucrative those opportunities are largely depends on where they sit in the value chain.

Elisa’s Sjöberg believes that “rather than treating satellite as a commodity or simple roaming add-on, telcos must offer outcome-based solutions – such as guaranteed uptime, actionable AI insights, and secure data sovereignty – to high-value B2B customers.”

That would enable CSPs to differentiate from enterprise satellite connectivity services already available on the market. In March, for example, US satellite service company MTN launched a “click-to-deploy” low earth orbit (LEO) satellite service on the Amazon Web Services (AWS) marketplace. MTN aims to provide enterprises with “last-mile connectivity and secure cloud backhauling to any remote site in the world” in as little as 24 hours.

But network operators will need the right skills and systems if they are to offer outcome-based solutions: “Monetizing integrated satellite and terrestrial connectivity hinges on telcos transforming from mere data providers to orchestrators within an evolving value chain," says Sjöberg.

Elisa has been taking part in the TM Forum Catalyst AI-powered supply chain and new revenue opportunities with satellite–mobile convergence - Phase III. The Catalyst team has been exploring how to unify satellite connectivity, mobile networks and intelligent edge capabilities to support real-time tracking, situational awareness and predictive decisioning for global logistics. This includes using AI agents to analyze multimodal data across air, land and sea to detect risks, anticipate delays and recommend optimization.

Work within the Catalyst has highlighted the importance of modern, flexible IT platforms that can support new ecosystem plays.

“Success depends on overcoming traditional traps like slow in-house development and outdated billing systems, and instead embracing plug-and-play integration that leverages their trusted relationships and national compliance, making the inherent complexity invisible to the end user,” according to Sjöberg.

Part of the Catalyst's aim is to use AI to automatically identify problems and provide solutions across different networks. Developing AI agents to securely analyze multimodal voice, text, video and sensor telemetry data across hybrid networks, however, is highly complex, points out Sjöberg. It requires “expertise in model compression for edge devices and temporal correlation to synchronize data across 5G towers and satellites”, he says.

UK cloud infrastructure company CGI, which took part in the Catalyst, points to the importance of considering the context when building AI agents.

"The newest set of AI tools ... are an orchestrator that has some specified tools at hand. It might be an LLM, it might not be an LLM; it depends on the topic," explains Hendrik Oppenberg, Executive Consultant, CGI.

"A good AI agent is a deterministic program that has access to AI for input/output. Having those really specialized on [a] container ... is more feasible than huge ...AI on prem somewhere," says Wolfgang Robben, Executive Consultant, Project Manager, Space, CGI.

"It's smaller AI that understands anomalies, for instance, in the temperature curve or in deviation from a projected route [or deviation] from SLAs. AI on its own, can take matters [in hand], or at least notify the operator."

At the same time. a critical game changer is the standardization of the "unified core", specifically through the adoption of 3GPP Release 17 (and beyond) standards for Non-Terrestrial Networks (NTN), explains Sjöberg.

“Without this specific technical collaboration, the seamless service layer ... for global logistics remains a collection of fragmented, expensive and incompatible silos,” he explains. “With this standardization, device manufacturers will join the movement.”