More and more, companies are reaping big rewards from their investment in the Internet of Things (IoT). Those who remain on the sidelines often ask when, or if, their business should jump into the game.
Some lack expertise, or are unsure about the technologies involved. But for many, the issue is mainly about the perceived investment of time. Making a case for IoT might seem like a moonshot when the question involves getting board approval for a project that might take six months, or could take three years.
To get the most ROI from an IoT initiative, we’ve found that building a predictable timeframe from initial investment to rapid completion makes sense to even the most cautious decision-makers. To help understand the process, here are five key phases we see customers move through in an IoT implementation, and how to get the most out of each:
No 1: Business case development
Typically the business case for IoT is handled by a cross-functional team and approved by business division executives or even the board of directors. It can be a fairly straightforward process, but companies often suffer from insufficient collaboration across the disciplines involved, and a lack of focus when it comes to potential benefits.
Trying to do too much can quickly become overwhelming, while disconnects between groups like engineering and maintenance teams can lead to difficulty in identifying which KPIs to target or an unrealistic view of how processes may benefit from more data and intelligence. Additionally, development of a business case generally falls outside of IT, and appropriate technical experts may not be brought into the room early enough to head off any dead ends or flag potential issues such as extensive coding required for integration.
To make this phase as successful as possible, customers tend to benefit most when they involve an executive champion early in the process and make sure a tight, focused set of KPIs is defined up front and validated by those responsible for carrying out the process.
To smooth things even further, consider bringing in experts such as those provided to eligible enterprise companies through Microsoft’s IoT QuickStart program. These are half-day sessions with established experts that can help the organization get oriented and begin implementing IoT concepts right away.
No. 2: Build vs. buy and vendor evaluations
After establishing what the solution will look like, most companies have a decision to make: Do we build it in-house, or find an external solution partner?
The answer is often some combination of both, with internal expertise, technology stack and cost all playing a significant role. While internal talent and cost are easier to evaluate, the technology itself can take more time to understand — What are the capabilities of various platforms and solutions, and how well do they integrate across the companies existing IT systems?
The last thing companies want is another fragmented solution requiring extensive integration. The whole point of IoT is to get data flowing. The best way to avoid the limits of a fragmented system is to choose technologies that work well together. Make sure and consider the technology stacks at any branch offices or other locations to ensure they’ll work as well.
Fortunately, this phase is also becoming much easier as the market matures. There is now a growing set of vendors creating packaged solutions and technologies for IoT. Microsoft’s offering is the recently available Azure IoT Suite, which provides key services needed to build an IoT solution from end to end, with the addition of pre-configured solutions for common scenarios. Now enterprises have an easy and seamless way to connect people, devices and assets that help them realize the opportunities of IoT without massive investments in infrastructure that would slow time to value.
No. 3: Proof of concept
Now that you have scoped a very detailed business case, it’s time for a proof of concept (PoC), which is designed to validate a few key points, not every single detail. The goal is to quickly assess feasibility—not to see whether you can do it all. For example, I have had many customers that have hundreds of scenarios and ideas they want to implement in their PoC. My advice has been to just start with the 1-5 scenarios or feature designs that matters the most to their business. Once we validate those first few, we can then evolve during the next (pilot) phase. We definitely want our customers to “think big,” but starting small during PoC enables them to experiment quickly and keep iterating.
If your executive champion is fully on board and has enough clout, hiccups in the process are much less likely to result in the plug being pulled. If your cross-disciplinary team is communicating and functioning well, most of the details should be well thought out.
In addition to the above, having a small, agile team that is supporting your PoC to quickly knock down barriers will also help. To get started on PoC quickly, businesses can use Azure IoT Suite’s preconfigured solutions, which are engineered to help company get started in minutes, instead of weeks or months.
No. 4: Initial pilot rollout
Once the concept is proven, it’s time to evolve the scenarios and make sure the IoT solution can be integrated into the broader organization. The biggest challenges at this stage involve training employees to use the system and preparing for any organizational changes the new process will require. The way many companies have approached this successfully is with a “limited” pilot rollout over a small geography. This will help vet out any unforeseen process gaps, identify new needs, and more before broad rollout in step 5.
From a technical standpoint, the biggest issue here (and in step 5) will likely be scalability. Can your system handle the increased volume? For most large organizations, the simplest way to ensure the right capacity is to plan ahead for growth and global reach by enlisting the help of cloud services such as Microsoft Azure, which runs on a worldwide network of Microsoft-managed datacenters across 22 regions—more countries and regions than Amazon Web Services and Google Cloud combined.
No. 5: Commercial deployment (full scale rollout)
The challenges at this final phase largely mirror those that typically crop up during the initial pilot rollout phase. Make sure all users of the system are trained and have bought into the benefits, and that all organizational changes and new processes have been implemented. Great internal training is a must to keep this process rolling smoothly.
At this point the solution will begin generating enormous amounts of data, and many customers struggle to understand what to do with it. Having the right expertise in place will ensure the company begins to gain the kinds of formerly hidden insights that make IoT implementations so valuable. Consider augmenting your IoT solution’s analytics capabilities by hiring a data scientist or a statistician early in the process to help drive advanced analytics.
Short of that, we advise all companies thinking about IoT to simply save their data. The data you’re generating today can become valuable later for historical analysis of trends, which can lead to predictive capabilities down the road. Cloud offerings like Azure can be valuable here too, ensuring that data can be securely stored until it is needed.
Moving through all five phases is an investment for any company, and our customers consistently report that they spent between 12 and 36 months on the process. But with new offerings like the QuickStart program, Azure IoT Suite and an emerging spectrum of packaged solutions from partners around the globe, that timeframe is being drastically shortened — down to a few weeks or months in some cases.
Visit us on the web to learn more about how to create the Internet of Your Things today.