As cellular IoT deployments grow from thousands to millions, the limits of hardware, not software, come into focus, writes IoT connectivity provider Onomondo. Until connectivity infrastructure evolves to adapt to constrained devices – rather than forcing them into legacy telecom models – cellular IoT will remain stuck in a cycle of complexity, fragmentation, and unrealised potential.
Too many people still think IoT scales like the internet. Like software. Like OS settings and apps on laptops, smartphones, and tablets. In reality, IoT – including cellular IoT – scales like hardware. Basic hardware. Whether you are scaling NB-IoT or LTE-M smart meters, asset trackers, e-scooters, trucks, or vending machines, you are never dealing with an iPad. You are dealing with resource-constrained hardware components, often in the thousands at once, scaling to millions.
Yes, you can tweak firmware over the air, but that is usually limited to patching bugs and security vulnerabilities. The underlying hardware – the ‘thing’ – does not change. Whether it is a smart meter sitting in a basement for 15 years or an asset tracker constantly moving across networks, you cannot reconfigure its network behavior in the same way you would an iPad. This is what creates a scaling constraint in cellular IoT.
It is also why the role of the mobile virtual network operator (MVNO) is to build connectivity infrastructure that can adapt around the product and application – so that even simple hardware can be made to behave more like software. Yet the industry still behaves as if the next generation of the cellular standard will solve the scaling problem – 5G, 6G, SGP.32, and whatever else was announced at Mobile World Congress in Barcelona.
But even 2G was sufficient for many IoT applications. The issue is not how fast the connectivity can go. The challenge lies in the connectivity infrastructure the technologies rely on. Because when IoT hardware is deployed in the real world, the connectivity layer becomes the operational backbone of the product or application. And today, that backbone is still too often built on fragmented, opaque telecom systems that were never designed for machines.
The result is a road full of potholes – reliability issues, security risks, unpredictable connectivity behavior, and operational complexity. Upgrading the vehicle will not fix that road. Most IoT applications send small amounts of data intermittently. They do not need Ferrari-level performance. What they need is connectivity infrastructure that is reliable, secure, and adaptable to the hardware constraints of the device.
The IoT equation is very simple: hardware + connectivity + cloud. In most deployments, the hardware and the cloud are relatively fixed. The only part of the equation that can truly adapt is the messy middle: connectivity. If that layer is programmable and transparent, then, as noted earlier, even simple hardware can be made to behave more like software. This is what will allow IoT deployments to scale globally across networks without rebuilding devices or renegotiating carrier relationships every time something changes.
This is where the role of the MVNO needs to evolve. For too long, many MVNOs have simply resold connectivity built on legacy telecom infrastructure. That may simplify procurement, but it does little to remove the operational complexity IoT deployments face at scale. Because the companies building connected products do not want to become telecom experts. They want devices that connect reliably and securely and work wherever their business operates.
They want outcomes, not infrastructure headaches. No one expects to walk out of their home 20 years from now and discover that cellular networks have disappeared. Cellular infrastructure will remain one of the most pervasive global communications systems ever built. For mission-critical IoT deployments, it remains the most realistic long-term foundation. Yet cellular IoT adoption has not scaled as quickly as many expected.
This has led some to question the future of the sector. In reality, the industry is simply moving through a familiar technology adoption cycle: early hype, followed by disillusionment, before long-term growth. Cellular IoT is not struggling to scale because cellular networks are unreliable or insecure. It is struggling because the MVNO industry has not yet fully embraced a simple idea: cellular connectivity must adapt to the application, not the other way around, if it is to stop being a bottleneck.
This shift clears the way for progress, and becomes the foundation that allows builders, engineers, and product teams to focus on solving real-world problems. Companies deploying cellular IoT at scale need their connectivity providers to act as infrastructure partners – helping them build, deploy, and operate globally without the friction that has historically slowed adoption.
It is this mindset that has helped companies like mine win global deployments with organizations such as shipping giant Maersk. Because scaling cellular IoT is not about buying a SIM card with a data plan and connecting a device. It is about building the frictionless infrastructure that allows millions of devices to operate reliably across borders and networks.
Once that foundation exists, cellular IoT can finally scale the way the industry always expected.
