The internet of things (IoT) represents huge growth potential for communications service providers (CSPs). At the same time, the IoT also poses a risk to the profit margins of connectivity providers that don’t strictly enforce IoT network service parameters.
At their most fundamental, IoT services deliver low-bandwidth, low-cost wireless connectivity to large volumes of distributed IoT sensors and devices. Turning a profit from basic IoT connectivity services presents its own challenges; it’s not uncommon for an IoT device to incur varying yearly network charges. So it’s imperative that CSPs keep their IoT service delivery costs as low as possible.
For providers to deliver continued connectivity at minimum cost, the IoT device must adhere to strict network usage rules that limit the number of packets the device transfers, the number of session connections per month, and other key attributes. CSPs use these parameters to establish the cost of providing the service and to determine the contractual price the customer must pay.
IoT devices that operate outside of the contractual agreements quickly eat into profit margins. Providers, then, face a tricky balancing act between closely monitoring the resource consumption of their low-margin IoT connections and managing the cost of doing so.
Delicate Balancing Act
Deploying complex monitoring equipment for IoT resource monitoring can be expensive. If factored into the fee that you charge customers, it might be more than customer business models can tolerate. The upshot is that you need a low-cost, automated way of regulating IoT connections that keeps devices compliant without adding prohibitive overhead costs.
A strong option is to employ simple data analytics to monitor key IoT metrics on a per-connection basis. In recent years, it has become increasingly cost-effective to deploy data collection and analytics tools that leverage telemetry from key network assets already used in daily network operations. The data processed by these systems can do double duty to add significant value and profitability to IoT.
For example, IoT devices are often highly distributed and deployed in areas not easily serviced. If a device no longer connects to the network, you need to quickly learn the reason. Is network coverage down? Is it a module problem? Is the IoT device’s battery dead?
These are seemingly simple and obvious questions; however, getting to the bottom of them could take days, weeks or months if you don’t have easy access to the multiple streams of data already stored in their networks that hold the answers.
Network Analytics For Automated Answers
To be a true IoT connectivity provider, you must be able to quickly troubleshoot issues and ensure that IoT devices are always connected and transmitting packets. Modern network analytics systems make this possible by economically collecting and analyzing data from the radio access network (RAN) and across the core. Once collected, this data is processed per session and per customer. The analysis is further enriched with device type, location, weather, and other data to provide true end-to-end visibility.
For example, let’s say a managed IoT service provider or large enterprise IoT customer deploys thousands of devices across multiple states and connects them to your network. After a few weeks, the IoT customer notices that several devices aren’t passing data to the servers. To find out what’s wrong, the customer will likely call you to investigate. Without network analytics, you might spend days or weeks pulling historical RAN, provisioning and session data from data warehouses for troubleshooting.
By contrast, with modern network analytics, each provisioned device would be automatically monitored and a baseline of normal behavior established. Monitored parameters would include destination IP addresses, module type (vendor/model), connection frequency and bandwidth consumed. Deviations from these baselines would generate proactive alerts to both you and the customer. A provisioned device that failed to connect would be considered outside of normal behavior, and additional data for root cause analysis would be reviewed.
RAN logs would be collated to determine if the device had attempted to connect to the network and if network congestion or coverage were the reason it failed to connect. Additionally, it would be possible to determine if this problem was a one-off issue or systemic in nature. If systemic, the commonalities among the failures would be quickly identified and provided to the IoT customer.
These network analytics techniques can both assist operations and help ensure contractual IoT compliance. Any deviation from contractually agreed-upon parameters is immediately detected and brought to the attention of operations teams. The device can then be disconnected, or the managed IoT provider or enterprise IoT customer can be notified to fix the problem.
Protecting Margins While Adding Value
Network analytics allow you to not only protect IoT profit margins, but also to become a true value-added partner. Beyond the basics of providing a low-cost way to ensure that IoT connectivity parameters are always in scope, analytics enable you to layer on other types of value-added services to further monetize IoT. Such services might involve IoT data storage and striking industry partnerships in vertical industries to support IoT for connected cars, smart cities, health monitoring, inventory management and other opportunities.
To get the most value from network analytics for IoT, consider tools that are low cost and leverageable across all business and technology domains, and that truly enable IoT operations. Without these tools, you risk profits and satisfied customers. You need to be able to demonstrate true value to the IoT operator — simply providing low-cost connectivity will not win over time. Providing troubleshooting support, security analytics and other insights will differentiate you from your competition and can help retain those accounts over the long run.
It’s easy to utilize legacy tools to operate the network. However, these tools can be cumbersome and expensive. Leveraging analytics has the potential to make the data and insights more valuable to network operations, planning, marketing and your end customers. Tools must create value for all key constituents in your organization, or they are simply a cost. Make your tools a revenue-generation resource.