For more than 15 years, the cost of cellular IoT connectivity has been falling. Prices per megabyte and per connection have declined steadily, making connectivity more accessible but also reducing the scope for providers to differentiate themselves on price alone.
That is changing the basis of competition. As connectivity becomes increasingly commoditised, enterprises are placing greater emphasis on technical capabilities, service quality and total cost of ownership (TCO). The emergence of SGP.32 eSIM technology adds to this pressure, giving customers greater flexibility to change connectivity providers remotely.
The result is a market in which the quality of the supporting technology and services are becoming a more significant differentiator alongside the headline price of a data plan.
Connectivity Management Platforms (CMPs) are a key part of the customer proposition. However, their capabilities vary considerably. Some focus on straightforward dashboards, reporting and bulk SIM management, while others provide more extensive access to raw data, APIs and integrations with enterprise systems.
More advanced platforms may interface to multiple network cores and support end-user billing arrangements and sub-accounts. Support for features such as multi-IMSI, Power Saving Mode (PSM), extended Discontinuous Reception (eDRX) and 5G RedCap can also be important, depending on the application.
For enterprises managing large, fragmented estates, CMP abstraction offers another option: a single interface across multiple connectivity providers and CMPs. This can simplify SIM lifecycle management, provisioning, analytics, security monitoring and billing. However, abstraction can limit some platform capabilities to the common functionality supported across the underlying platforms.
Connectivity abstraction goes further, managing devices and assets rather than just cellular connections. It can unify management of cellular, satellite, Wi-Fi and fixed-line connectivity, streamlining management of device security, firmware, applications and access controls. This is particularly relevant as IoT deployments increasingly combine different communications technologies and edge computing.
The GSMA's SGP.32 specification is designed to make eSIM remote provisioning more practical for constrained IoT devices. It uses an IoT Profile Assistant (IPA) on the device, controlled by an eSIM IoT Remote Manager (eIM) to download and manage profiles without requiring human intervention.
This allows devices to be manufactured with one connectivity profile and provisioned for local networks after deployment. It also creates the potential to switch providers remotely, whether to improve coverage, service quality or price.
Large-scale migrations between providers may remain relatively uncommon, but the ability to switch creates a competitive pressure. Effectively, providers will need to demonstrate that their networks and services offer compelling reasons for customers to stay. Proprietary extensions to SGP.32 may, however, limit interoperability between some implementations.
A dedicated mobile core network can give a connectivity provider greater control over service delivery, even when using radio access networks operated by partner mobile network operators. It can help standardise network behaviour, fault escalation, service levels and upgrades, while providing better visibility of performance.
Local traffic breakout (LBO) is another capability that can make a material difference for particular applications that might operate in multiple geographies. Rather than routing all traffic back through a central home network, data can be routed locally through a local gateway.
This can reduce latency, network loading and costs, while helping address data sovereignty and regulatory requirements. It is potentially valuable for connected vehicles, for example, where software and firmware updates can generate substantial volumes of traffic. Local breakout can also support edge computing and corporate private network deployments.
Ownership of software intellectual property can give providers greater control over their services, improve agility and reduce licensing costs. This can be especially relevant for high-volume applications with relatively low revenue per connection. However, developing and maintaining proprietary software capabilities requires significant investment and scale, so third-party solutions may remain appropriate. Or some hybrid approach.
Artificial intelligence is also becoming increasingly relevant to network operations. Potential applications include predicting congestion and failures, optimising network performance, improving energy efficiency and automating troubleshooting. AI can also help identify unusual device behaviour and trigger appropriate responses, including reconfiguration or isolation.
Security is an increasingly important part of the proposition. Providers may offer AI-powered network anomaly and threat detection for traffic-flow analysis to identify suspicious behaviour, as deep packet inspection can introduce privacy, regulatory and commercial complications. Additionally private APNs, access controls, VPNs and SIM-based security capabilities such as IoT SAFE may be supported.
Enterprises should also examine relevant security and resilience credentials. These may include ISO 27001, SOC 2 Type II, ISO 22301, IEC 62443 and GSMA security frameworks, depending on the services involved. Redundancy, failover, backups and recovery arrangements are equally important in avoiding single points of failure across networks, platforms and data centres.
There is no universally appropriate set of technical capabilities for a cellular communications service provider. Local breakout may be essential for some automotive applications but unnecessary for a low-bandwidth sensor. Voice support may matter for emergency communications and assisted living, but may add little value elsewhere. Connectivity abstraction becomes more relevant when deployments combine cellular with satellite, Wi-Fi or other technologies.
The challenge for enterprises is therefore to assess providers against the technical and operational requirements of their applications, rather than focusing primarily on connectivity prices.
As the costs of raw cellular IoT connectivity continue to fall, differentiation will depend more on the platforms, software, security, resilience and network capabilities that surround it. Price will remain important, but it is becoming only one aspect of a much broader purchasing decision.
Learn more
This theme of CSPs finding new mechanisms for differentiation and monetisation is further explored in the recent Transforma Insights report 'Beyond price: procuring cellular connectivity for IoT'
