Today, the promise of IoT is in little doubt. Use cases range from connected products to connected assets and we are seeing IoT deployed across a multitude of industries from telehealth and electric vehicle (EV) connectivity to smart vending, payment systems and more. Accelerated in part by the pandemic and remote working, IoT deployments are now becoming mainstream.
That said, deploying one IoT device or a prototype and ensuring it is functional, versus deploying at scale across multiple sites and geographies is when IoT starts to become more challenging.
In fact, it is estimated that as many as 75 percent of all IoT projects either fail or don’t reach their full potential and deliver the expected outcomes. According to our State of IoT Adoption Study, undertaken last year where we surveyed 500 UK and USA-based senior decision-makers, 77 percent of respondents said their IoT project was at best only somewhat successful in meeting expectations and realizing benefit.
Successful IoT deployments involve solving a wide range of challenges at the device level and then scaling that device, so it works everywhere. This starts at the initial concept and design to provisioning to deployment; to scaling up to wider implementation; to ensuring the device is secure. This is in addition to the ongoing maintenance and management of devices over their lifetime, once in the field.
Connectivity is the hardest element to get right
However, putting all these issues to one side, the hardest element to get right is connectivity at scale. For everything to work, the device needs to connect everywhere and every time so it can reliably transmit data back and forth. The last thing organizations need is to be stranded without communications, having invested heavily in applications, storage, security and operations.
Herein lies undoubtedly the biggest challenge. Enterprises often underestimate the complexity of reliable connectivity at scale. Many projects run into problems because they worked as a prototype but simply can’t scale once in the field. This is often because the nature of the physical environment, where devices will be required to work within, gets overlooked. It is therefore critical that a range of environments and use cases are considered at the outset.
All too often customers will rely on one network provider for connectivity. All networks have blackspots and geographic borders. Therefore, national, and moreover, global deployments need to be able to localize and/or roam onto different networks. In other words, they need network redundancy. Ideally, your IoT device should have the flexibility to connect to the best available network wherever a cellular network or public private option works at scale, and this requires connectivity to multiple network providers.
There is also a lack of planning for device or connectivity problems. Devices and networks fail, or become obsolete as spectrum is “refarmed” for next-generation consumer devices. Customers need to consider all likely scenarios and plan thoroughly for these.
Our research found security, connectivity and device onboarding were cited as top challenges; 39 percent said security was a big hurdle, while 35 percent stated device onboarding, testing and certification, and cellular connectivity across multiple countries and regions had proved difficult. Therefore, here are six key considerations that SIs need to consider in global deployments.
1. Make the right connectivity choices for the future
There are a wide range of connectivity options to consider, and it is critical that SIs think through all the possible places where a client’s thousands of IoT devices might end up — not just now but in the future. Today’s choices include Cellular, Private LTE/ 5G, Wi-Fi, LoRaWAN, Bluetooth, Thread and Satellite and other wireless options.
2. Make sure the solution can dynamically optimise to networks
Devices need to be able to localize to the best network wherever that may be. For example, deploying thousands of devices on a single mobile network can mean 15-25 percent are without a reliable connection. One solution is to select an IoT connectivity partner that supports eSIM for remote profile download and management onto the SIM as well as the eUICC. This manages the secure element within the eSIM and enables the hosting and management of multiple global network profiles, allowing devices to dynamically access and choose the strongest and most reliable network.
3. Get the hardware right
Connectivity relies on various software and hardware elements being properly configured and working in tandem with the SIM and modem. Different networks have varying certification requirements and devices will have differing size and power considerations. Therefore, all hardware and firmware needs to be designed with the variety of available options in mind.
4. Test and validate
When networks and connections fail devices need to be able to fail and recover quickly. A well-designed device should be able to spot when a network is down and connect to another or implement a backup protocol until it can reconnect. The best way to spot potential problems is to complete thorough testing. This means observing device behaviour, verifying how it connects, and running resilience tests in the lab to understand how the device responds to unexpected events which could face it out in the field.
5. Plan for lifetime management
When planning customer deployments, consider connectivity over the lifetime of the device. Undertake a total cost of ownership exercise that looks beyond the initial deployment to how the implementation might scale in years to come. In our experience, underestimating the total cost of ownership is common, so consider all future scenarios, including selecting appropriate primary and backup connectivity technologies for devices.
6. Ensure IoT connects through platform-enabled services
Devices need to integrate seamlessly into your client’s environment, therefore managing an IoT estate needs centralized platforms which give oversight of IoT data and allow for device management. They may need to transmit data into your environment via private networks or cellular or they may even need to switch between the two. This means getting the software right on both the device and the network is critical so the two can communicate wherever they are.
Partnering with an IoT connectivity expert
Successful IoT deployments are not for the faint-hearted. SIs need to ensure they have the right IoT connectivity partner alongside them to help navigate through the complexities associated with IoT deployments, and free them up to focus on all the other aspects of deploying IoT at scale.
The market opportunity for IoT is enormous. In our survey, 86 percent of respondents said IoT is a key priority for their business. However, increasing device connectivity is key to unlocking value and innovation in global IoT deployments. Therefore, businesses need to consider their connectivity requirements at the very beginning of any IoT project, working with IoT connectivity experts to ensure this is factored into device design, manufacture, and testing phases.
Image credit: shawn_hempel/depositphotos.com
Mike Cihra is SVP Channels and Alliances, Eseye