Satellite connectivity flies but enterprise integration remains chaotic

Where satellite already functions as core infrastructure

Satellite connectivity still acts as a resilience or backup layer in many environments. However, a marked shift has emerged in the past few years. Sectors such as aviation, maritime, energy and defence are now embedding satellite as part of their core network infrastructure.

Sometimes this is as the default WAN transport, rather than an afterthought. For highly remote operations such as some construction and mining sites, which may not have a lot of terrestrial infrastructure, it is also the main connectivity option. Temporary infrastructure in disaster zones as well as some types of large-scale events may rely heavily on satellite too. In many of these use cases, it is integrated into hybrid SD-WAN architectures.

On the other hand, in more traditional enterprise networking situations, the technology is still largely considered optional and is usually deployed as part of a hybrid resilience strategy.

“Satellite is extremely valuable for disaster recovery, remote operations and continuity planning, but the management layer has to match the criticality of the use case,” suggests Tim Last, executive vice-president at Iridium. “The most effective deployments are designed around the application, not just the connection. That means defining what must stay online, what data needs priority and how the network should behave when terrestrial infrastructure is degraded or unavailable.”

As such, outside of mission-critical situations, broader satellite adoption is still determined by environment, geography and type of operation.  

The hybrid integration problem

The majority of the challenges organisations deal with when it comes to satellite integration stem from the ground realities of merging vastly different network environments. Conventional enterprise architectures were mainly designed for terrestrial infrastructure such as broadband, fibre and cellular networks.

“These environments generally provide relatively stable latency, deterministic routing behaviour and mature operational standards,” explains Khaled Elbehiery, professor at OPIT – Open Institute of Technology. “In contrast, satellite systems, particularly modern LEO architectures, introduce highly dynamic network conditions due to continuous satellite movement, beam handoffs, gateway transitions and changing orbital topologies. Enterprise applications and networking tools were simply not designed with these characteristics in mind.”

One of the biggest technical challenges are dynamic and high mobility handovers because of fast topology changes. Due to LEO satellites orbiting relatively close to Earth, they usually move at very high speeds reaching up to 17,500 mph. This means that a specific ground terminal can realistically view a satellite for only around five to 15 minutes usually, prompting very rapid handovers from one satellite to another. 

For organisations using real-time applications such as financial trading or video calls, these handovers can cause micro-interruptions and packet loss in many cases. 

Protocol and routing incompatibilities have significantly complicated satellite integration into hybrid enterprise networks too. LEO propagation delays can vary significantly, which can lead to algorithms controlling TCP congestion to wrongly assume that delays mean packet loss or trigger false retransmissions. 

“In many cases, LEO systems can achieve latency levels comparable to terrestrial broadband. The real challenge is latency variability,” OPIT’s Elbehiery adds. “Applications typically tolerate higher latency more effectively than unpredictable latency fluctuations. A stable 80-millisecond connection often performs better operationally than a connection fluctuating between 30 and 150 milliseconds.”

Routing paths for LEO networks also change extremely fast, often needing SDN routing in space to maintain session state. In contrast, terrestrial routers usually use slowly changing OSPF/BGP updates or static tables. Another integration hurdle arises due to organisations depending on rigid service-level agreements (SLAs) for jitter, availability and throughput for end-to-end predictability.

However, replicating this in a hybrid terrestrial-space network environment means having to standardise how air interface from a satellite translates and connects to core terrestrial networks. This would let satellite signals use the same billing, routing and management systems as terrestrial infrastructure.

Satellite is extremely valuable for disaster recovery, remote operations and continuity planning, but the management layer has to match the criticality of the use case

Failover orchestration is another major hurdle. Despite satellite being increasingly positioned as a resilience layer, seamless traffic transition between satellite and terrestrial networks without affecting applications remains significantly challenging. As such, enterprises have to be constantly mindful of consistent security policies and routing stability to prevent any service interruptions during transitions.

“Failover orchestration represents perhaps the most difficult operational challenge facing enterprises today. The issue extends far beyond simply switching traffic from one path to another,” Elbehiery says. “Enterprises must maintain application sessions, preserve routing stability, synchronise security policies, avoid overlay tunnel resets and prevent route oscillation during failover events. Many organisations experience ‘micro-outages’ or temporary application degradation during transitions even when connectivity itself technically remains available.”

Similarly, cloud integration remains complicated, since modern cloud platforms were also built around the same terrestrial networking assumptions such as predictable ingress paths and stable latency. This leads to organisations often experiencing fluctuating round-trip times and variable jitter when they try to bring satellite into the equation, impacting everything from virtual private network (VPN) stability to cloud application performance. 

Crucially, despite satellite networks behaving like a completely different medium, most organisations are still left trying to shoehorn them into systems designed primarily for land-based use. 

Visibility and control gaps across hybrid networks 

Even after satellite is integrated into hybrid networks, visibility can continue to be a significant challenge. In many cases, satellite links introduce the most volatility to networks while also being the hardest to consistently observe due to a mix of orbit dynamics, beam transitions, variable radio frequency conditions and gateway handoffs.

One of the biggest operational blind spots across hybrid networks is that there is often no unified telemetry across different layers, with data spread across multiple dashboards. Most hybrid setups have a satellite operator, a cloud provider, an SD-WAN provider and a terrestrial ISP, all with their own telemetry, SLA metrics and logs. 

As such, teams can struggle to pinpoint whether it’s a satellite, cloud ingress, SD-WAN routing or local ISP congestion issue during performance issues. 

This can lead to significant root-cause analysis confusion and longer mean time to recovery (MTTR), especially during latency spikes, connection drops and outages. As such, troubleshooting often becomes far more of a cross-domain guessing game than a confident remedial strategy.

Other business and governance blind spots can emerge too. Without clear data ownership, enterprises are also open to a number of regulatory and compliance risks, especially for government and mission-critical data, as well as more complicated and expensive audit procedures.

Similarly, enterprises may have to navigate security considerations, with escalating cyber security risk because of the “satellite encryption gap” too. This is due to a large amount of satellite-related data like voice calls and internet traffic transiting to ground stations inconsistently unencrypted, making it more vulnerable to eavesdropping.

Highly siloed telemetry across ground and space segments means that security teams can often not validate the constant behavioural analysis and micro-segmentation required by modern security protocols, leading to zero-trust failures.Rising cloud egress costs due to applications sometimes needing unnecessary information transfers and higher data transfer fees because of a lack of streamlined views can also pose cost challenges.

Why satellite adoption still lags

Even in cases where satellite integration shows clear benefits, it often struggles to transition from isolated deployments into core scalable enterprise infrastructure. This is mainly because most organisational and supplier structures are not designed around satellite as a core transport layer.

Enterprise-grade, multi-orbit satellite terminals – which need to balance connectivity across terrestrial fibre, LEO satellite links and cellular networks – are still very expensive. In many cases, they also need specialised power facilities.

As a result, integration costs often scale across the whole network stack, with satellite becoming a marginal addition that can greatly increase operational complexity expenses.

In many cases, organisational fragmentation itself lies at the core of slow satellite adoption. This is because in several well-established enterprises, cloud, network and security teams still don’t align fully.

With existing teams already dealing with highly siloed data and slow-moving processes, hybrid connectivity often has no single owner. As such, adding another layer of complexity, especially one as highly volatile and difficult to observe as satellite, can automatically take lower priority.

A lack of industry-wide operational standardisation is another very common obstacle contributing to enterprise hesitancy around satellite adoption. Different providers usually have a variety of control planes, with no internationally well-established shared operating model across satellite, terrestrial and cloud yet available.

However, in many cases, enterprise inertia and overall risk-averseness plays a much bigger role in slow satellite adoption. Many organisations still greatly avoid changing stable architectures unless absolutely forced to and are content to stick with “good-enough” existing SD-WAN and terrestrial setups. 

Towards seamless satellite integration

For satellites to move beyond one-off deployments towards core infrastructure, enterprises need to rethink their fundamental operating models. One of the biggest steps towards seamless integration would be establishing a hybrid networking strategy across satellite, terrestrial and cloud, which includes standardising telemetry and routing visibility models. Developing stronger supplier coordination across network layers is also essential.

“The industry needs more interoperability, better abstraction and stronger operational integration. Satellite should be easier to consume through the same tools enterprises already use for cloud, SD-WAN, security and network performance management,” says Last. “Standards will help, but so will practical supplier collaboration around APIs, telemetry, automation and service assurance. The future is not satellite replacing terrestrial networks. It is satellite becoming a normalised, trusted part of hybrid enterprise connectivity.”

As such, enterprises who adapt their operating systems to integrate satellite as core infrastructure, alongside terrestrial and cloud will be in a much better position to build resilient global connectivity down the line.