MWC crowd

This week in Barcelona there was a lot of excitement about networks designed to support the Internet of Things (IoT) and Smart Cities. They included several ecosystems of low-power wide-area (LPWA) networks that run in unlicensed spectrum, and emerging options in licenced spectrum from the traditional cellular players.

Such excitement is understandable. Forecasts predict explosive growth in device numbers over the next five years, and shows like this are a great way to see what’s possible at the bleeding edge. But as usual at MWC, the economics of these new services were hard to find amongst the slick technology demos.

Given the diversity of technology options to support wireless IoT, and that the business case metrics are fundamentally different to consumer mobile, it is surprising that the business models for IoT networks are so similar, and so conventional.

The unlicensed LPWA operators seem to be replicating a conventional cellular operator model. Build a network for national coverage then generate connection- and usage-based revenues. It's a high stakes strategy for LPWA operators because it relies on massive volumes of very low-value machine transactions to make sense.

Cellular operators have the opposite problem. Even with the massive device growth predicted, tiny per-device revenues mean that IoT is likely to remain a tiny fraction of their business. Yet operators will have to make long-term service level commitments (a typical IoT service lifecycle is 15 years) on this marginal business line.

The conventional model for cellular networks was “build it and they will come”. But if you swap the proven reliability and sheer size of consumer voice and data revenues for the smaller unit size and unproven revenues of IoT, then it doesn’t look like such a good fit.

Perhaps there is another model emerging. We know that the typical business case for smart city applications is citywide, not nationwide, with service ecosystems connecting and automating between local services and infrastructure. It points to different network thinking: local coverage, local cost and service control, local service innovation.

A use case that is pretty universal (which also has the great momentum of a growing wave LED conversions behind it) is the wireless control of street lighting. The city or lighting operator buys the control system because it quickly pays for itself in energy and maintenance savings, and as a bonus the city gets its own local LPWA network.

One million streetlights later, we can attest that this is a business model that works. Cities move surprisingly quickly when they are investing to save not burn taxpayers money, enabling sensors and monetising applications.  Cities are also a motivated driving force in joining together other local services, such as social care and transportation. This last part is in its infancy, but looks very promising indeed.