For many businesses, IoT has rapidly become a powerful “weapon of mass disruption.” Regardless of whether your goal is to simply track data and control devices, or you aim to combine IoT with big data, artificial intelligence (AI), and machine learning to create a truly connected enterprise and transform your business model, you’re likely to encounter challenges. Complexity. Codependencies. Legacy architecture. Interoperability challenges. What could go wrong? It might not surprise you to learn that, just 26% of all IoT initiatives succeed.
Here are four strategies you can adopt to greatly increase the odds of success for your IoT or IoT-enabled connected enterprise project.
1. Start small
IoT solutions are complex and involve a wide range of capabilities, some of which may be new to your organization and hard to find: cloud, analytics, integration, process changes, AI, machine learning, and more.
To counter what might seem like a big hurdle, we recommend that you start with a small project. Find a discrete problem or opportunity that can be easily addressed with an IoT solution. For example, you might start with a cloud storage or mobile payment solution that is small enough to complete successfully in a relatively short amount of time. Small projects help you build the skills you need for bigger, more complex projects, like mobile apps and smart meters.
2. Plan to succeed—and to scale
Think of your solution from the beginning as a system that will scale. Start by considering these questions. Will you use a reference architecture? How will you connect all the devices? Which IoT middleware platform will you use? How will you build in security and robust testing?
To reference architecture or not
A reference architecture provides a predefined set of architectural best practices for your solution. In many ways, it can act as an accelerator for your project.
If your IoT project is small, and limited in scope, you may not need one. If your IoT project is big and complex, a reference architecture can help you address things such as: connectivity and communications; device management; data collection, analysis, and actuation; security; integration with other systems; even disaster recovery.
Current general purpose IoT reference architectures include: IoT-A, Industrial Internet Reference Architecture (IIRA), and P2413 (IEEE Draft IEEE Approved Draft Standard for an Architectural Framework for the Internet of Things (IoT).
Decide how you will connect the sensors/devices
How you connect the “things” in your IoT solution will depend on many considerations. How many nodes? What capabilities and constraints do the nodes have (such as power)? What capabilities are already available (such as Wi-Fi)? What range of communications is supported by the nodes? What are the existing and future security and compliance requirements?
Once you answer these questions, you can select the most appropriate communications protocol: LPWAN, cellular, satellite, Wi-Fi, and/or Bluetooth.
You might, for example, choose an LPWAN-based system if your solution covers a wide geographical area. If you have many nodes that are power constrained, you might choose a lighter weight UDP-based protocol to put nodes to sleep and then wake them up on interrupt.
Choose an IoT middleware platform
The IoT middleware platform is where all the ingestion, processing, storage, and integration of data happens. To evaluate a platform, ask these questions. How scalable do you need your solution to be? How scalable is the platform? What platform bandwidth constraints exist? How well does the platform meet your security and interoperability requirements?
Current IoT middleware options range from open source (Kaa, SiteWhere, and others) to cloud (Azure IoT, AWS IoT, Watson IoT, and others) to hybrid on-prem/cloud (SAP’s IoT Foundations) to specialized (ThingWorx and others).
Build security in from the start, and from the edge
Whether you’re building a simple IoT solution or architecting a robust connected enterprise, you will likely be dealing with an ever-expanding network of millions or more endpoints and mobile users. Depending on your industry and location, you may have regulatory and compliance considerations. At the very least, your solution should include pervasive security that is built in from the very beginning, to the edge of the network. It should also include security for data at rest and in motion.
Account for QA
The complexity of all the interconnected devices and processes of a connected enterprise means that changing or adding one small thing in the system, or part of the system, may break the entire system. Build in robust, scalable quality assurance and testing procedures from the start so you can minimize unintended disruptions.
3. Work the APIs to unlock value
The apps you build using IoT APIs are the key to how you unlock real business value from the devices and data in your IoT solution.
The device APIs may be well defined and designed and you will find it straightforward to work with them. Some APIs are open source and will allow you to mix and match with systems you already have; some are not, and you will have to adapt the API’s language to your environment. In some cases, especially with older devices, you may not even have an API, and you will need to write one that can abstract the data from the device.
4. Partner to go faster
To plan, design, implement, and maintain an IoT system can be complex, requiring many skills that may be hard to find. Many companies that have completed successful IoT projects have found that working with outside partners provided the skills and expertise they needed to ensure their success.
Among companies with successful IoT projects, 60% engaged outside vendors in strategic planning; among manufacturers with successful IoT projects, 63% worked with partners during implementation.
Things to look for when considering an outside company to partner with on your IoT or connected enterprise solution:
• Proven experience designing, developing, and deploying IoT solutions
• Dedicated developer teams who can dig deep into your business and environment to help you build the right solution for your needs
• Deep skills in cloud-based infrastructure development, big data, AI, machine learning, and data analytics
• Expertise in UI/UX design and development
• Responsive, Agile approach so you can move quickly and cost effectively
What’s next on your journey to the connected enterprise?
If you’re planning an IoT solution or architecting a connected enterprise, let’s talk about how we can help you build faster!