IoT has far-reaching potential – if it is done correctly. Unfortunately, too often enterprises face serious challenges in their IoT projects and deployments, for a variety of reasons.
Sixty percent of organizations have failed to effectively take their industrial IoT initiatives beyond the proof-of-concept stage or beyond implementation at a limited number of sites1.
There are many reasons for this, but here are 10 steps that businesses can take to implement a successful IoT strategy:
1. Develop a clear business case. Cap Gemini found that about half of organizations struggle when it comes to establishing a business case for their IoT investments1. Especially when embarking upon IoT projects for the first time, enterprises may not appreciate the full promise of IoT for unlocking new possibilities. It is important to develop a sound and realistic business case rather than yield to market pressure to simply keep up with technology. Working with experienced IoT providers can help you develop a sound business case.
2. Be clear about what to measure. Many companies find it difficult initially to define the parameters to be measured by the IoT system while in the planning stage. When that isn’t done adequately, it hinders the effort to choose the most appropriate sensors and analytics systems going forward.
3. Understand the ROI. A major IoT project may have a very long timescale before realizing business benefits, which can lead a company to conclude that its project failed to deliver the anticipated return. But maybe they haven’t waited quite long enough. It requires skill and insight to know whether to extend a project a bit longer or to cut it off before it drains time and resources any further. This reinforces the value of starting small with a project and testing it thoroughly before expanding and working with a trusted IoT partner that has experience building and scaling IoT programs.
4. Design a sound system architecture. Technical design, according to the British Computer Society, accounts for as much as one-third of project failures2. The fault lies in the difficulty of translating from logical to physical design and the use of theoretical methodologies that focus on structures and relationships between engineered components. IoT deployments will, over time, need to accommodate evolving system architectures.
5. Secure top-level buy-in. Executive sponsorship and stakeholder engagement are vital for well-managed, well-led projects. Front-line managers don’t always have the scope of responsibility to see projects through to a rewarding conclusion, which is where the higher-level involvement can provide the necessary push. Over a long project lifecycle, it’s important to maintain that high-degree of leadership in the face of changes and shifting budget resources.
6. Grasp the full project scope and timescale. Project management involves dealing with a complex mix of interlocking interrelationships, strategic alliances, networks and contracts. If those contracts and their deliverables aren’t managed well through the project lifecycle, it results in damaging cost and time overruns. IoT projects can be more challenging in scope and complexity compared with other IT projects, requiring more complex project management skills.
7. Scale from pilot to full deployment. Small systems don’t always scale smoothly to large systems, and any bottlenecks in systems development processes need to be understood before moving from prototype to large-scale systems. Extensive testing is a must, including field trials to help determine potential issues that could be anything from a sensor interface to a faulty database in the cloud.
8. Understand the complexity. The complexity of IoT solution components is incredibly challenging to understand, especially when it comes to geographically broad initiatives with sensors in many different locales. The best technology for 100 sensors may not be the best for 100,000 of them. Choosing the type and size of remote devices, considering their power requirements, compute capabilities, subscription rates, and of course the connectivity options –will change and evolve over time – requiring patience and ongoing reconfiguration of your IoT plan.
9. Ensure adequate condition monitoring of the networks and sensor pool. IoT networks require frequent over-the-air updating of their many hundreds and thousands of remote devices. Detecting when a connection has failed requires good asset management in order to identify and fix problems. IoT is designed to precisely monitor a company’s assets, but when links aren’t operating properly, that precision is lost.
10. Bulk up your security. IoT devices are designed for simplicity, low power consumption, and a long life, not always for security. Devices with limited cryptographic capability and operating systems and little memory can be weak links, subject to hacking that compromises data and privacy. A company’s IoT network, with all those endpoints, presents a far broader attack surface, requiring close attention to security.
Clearly, there are many challenges when it comes to an enterprise’s IoT initiative. That points to the importance of the right provider, one who can partner and provide a secure, reliable, and potent network as well as work closely with an enterprise to overcome the challenges that it encounters in its IoT journey.
Sprint’s Curiosity IoT is designed to enable enterprises to fully manage their IoT devices and connectivity, through a dedicated, distributed and virtualized core network and operating system. The ultimate in security is provided on Curiosity, from the chip to the cloud, along with a vast network of vendors who can provide the elements needed for a successful IoT deployment and ongoing high performance. And, for those interested in a real-world testing environment, they can explore Curiosity Lab at Peachtree Corners where they can evaluate a myriad of IoT applications.
1Cap Gemini survey cited in a new Beecham Research white paper
2Why IoT Projects Fail white paper by Beecham Research, http://www.beechamresearch.com/