Anyone working in the roaming industry has probably realised two characteristics of the Internet of Things (IoT) from day one. Firstly, IoT is a global business, where connected devices are manufactured in one country, connectivity is delivered by a service provider in another country and the devices are tracked to and used in multiple other countries. Secondly, IoT roaming is completely different from consumer roaming. In the consumer segment, the traffic is seasonal, relatively predictable and the focus is on incentivising usage with attractive retail plans and campaigns targeted to silent roamers. Wheras IoT roaming is unpredictable as it depends on the IoT sales in the enterprise market, and it has different requirements in areas such as scalability, security, coverage, quality and latency.
Optimising the network edge for IoT
Roaming managers have quickly realised that their existing roaming agreements are not designed for IoT. Managing consumer and IoT traffic in the same wholesale agreements, discount models and billing processes are extremely difficult tasks and are sometimes even counter-productive. As an example, one operator may be a net receiver of consumer roaming but a net sender and payer for the IoT traffic.
Another example is the difference in the ability to steer devices to preferred networks compared to steering consumers. In IoT, operators need to be able to control the cost of outbound IoT devices and steer devices to the optimum networks, however the cost model for IoT may be different to the consumer wholesale roaming tariffs. Hence, there is a need to offer non-steered roaming in remote or mission-critical IoT applications. To make things even more challenging, while roaming agreements are usually negotiated annually between operators, in the IoT business model, enterprises require multi-year contracts with fixed data price plans.
IoT detection is fundamental
Operators are developing IoT services that go beyond connectivity to avoid being a ‘dumb pipe’ where other players innovate and manage profit models around the services and data from the IoT devices. Then again, roaming comes into the picture. An operator may end up losing from this connectivity service to inbound roaming IoT devices if this traffic is not properly monitored, measured and monetised. Indeed, mobile operators are also enablers for other operators by providing the essential connectivity and networks via their roaming agreements. For example, when bidding for multi-million smart meter contracts, operators have to guarantee the highest standards of coverage and may rely on international roaming agreements to use multiple networks within the same country. The roaming operator has, in this case, a competitive advantage over national operators due to the absence of national roaming. In such permanent roaming situations, specific commercial strategies, charging models and management tools are needed on the operators’ side.
The first step towards monetisation of IoT devices requires the ability to detect these devices on the operators’ network. This may seem obvious, but transparency principles defined in the mobile operators’ community for the identification of IoT traffic are not sufficient. The reality is that without efficient detection measures in place, the majority of the IoT traffic is not differentiated from the consumer traffic and silent IoT devices are undetected. Operators need to detect IoT devices, apply commercially agreed permanent roaming rules and segment based on different parameters like usage groups and radio access technology to apply the appropriate charging models.
IoT governance: how to deal with the compliance and security challenges
Thanks to software-based IoT detection and the application of permanent roaming rules agreed with roaming partners, operators can then monetise the inbound roaming IoT traffic. So everything starts with the capability to accurately detect the IoT devices and their permanent roaming status. Due to silent devices and technical challenges, an advanced detection solution requires presence discovery using signalling information and artificial intelligence to gain visibility and insights on their roaming traffic patterns. Alongside detection, operators will need to negotiate the right wholesale agreements with their roaming partners and have the proper, segmented, and device-based rating and settlement processes in place.
The settlement of the “standard” roaming traffic is already quite complex, with the need for dedicated software solutions to forecast traffic, simulate tariffs and track and reconcile discount deals agreed between partners. The addition of traffic segmentation and new charging models specific to the IoT business will make the use of such software solutions inevitable. In upgrading such solutions, operators should also look at ways to automate the reconciliation and settlement of the IoT roaming charges.
NB-IoT and the importance of being first
Narrow-Band IoT (NB-IoT), together with LTE-M, is intended to deliver on the promise of scalable, cost-effective massive IoT applications. Some of the main advantages of NB-IoT are lower power consumption, improved indoor and extended coverage and scalability. According to the latest Ericsson Mobility Report, at the end of 2025 NB-IoT and Cat-M are projected to account for 52% of all cellular IoT connections.
As mentioned earlier, IoT service contracts are long-term, and the enterprises are locked with an operator platform and services for several years. Therefore, it is important to be among the first operators to offer international NB-IoT coverage and deploy the required NB-IoT agreements. Here again, wholesale billing for the roaming usage of massive IoT will have to be different to the consumer roaming traffic and that is why new billing standards are also being defined, in particular the GSMA’s Billing Charging Evolution.