There’s a subtle reality that often gets overlooked during the ideation phase of an Internet of Things solution. The “Things” are real, physical products that have to be manufactured, shipped, installed, connected, serviced, billed, etc. Enterprise workflow supporting all of those tasks is critical to commercializing IoT products. These workflow and logistics issues are as important as the product itself and can become showstoppers if they are an afterthought.
It seems the mobile app industry has set an expectation that scale can happen practically overnight. While there is a great deal of innovation and technology that IoT is leveraging from the app industry, there’s one fundamental difference. Software-only apps (like your favorite puzzle game) leverage a hardware platform that is already deployed. Mobile app-only developers are fundamentally decoupled from the manufacturing, distribution and lifecycle of the supporting hardware platform. However, in the case of an IoT product, the hardware is often the manufacturer’s physical product, and deploying it is a huge part of commercialization. The point here is that the simple physicality of IoT requires careful consideration from the very beginning of the design process. One could argue it warrants as much time and consideration as the core features of the product itself.
Think through what’s required to ship a physical product. Your “Thing” starts as raw materials which must be manufactured, assembled, provisioned, tested, boxed, shipped, unboxed, installed and activated. Most of those steps require human interaction of some sort, a far cry from clicking “download” in an app store. Not to mention, this is just initially deploying the product. If the product has an issue once in the field, you might have to update it, replace it or remove it — which may require additional human intervention. Although, sending a service technician isn’t always a practical option once the device is out in the field or deployed at high scale.
Time is one of the key considerations. On the front side, you have to consider the time required to simply procure parts. Availability can be volatile, especially for the latest and greatest tech. You may have designed the most innovative and cost effective system, but it doesn’t matter if you can’t fill your bill of materials because the key parts are in high demand.
Cost is probably an obvious consideration, but its sources not always as obvious. Pennies matter at scale. Minor decisions in one area can have a major impact in other areas. For example, you may save both time and labor costs by purchasing components that are pre-certified or pre-tested, allowing you to omit tasks during manufacturing. That can add up to a major impact downstream when you multiply small time and cost savings a few hundred thousand times.
One of the most elusive but, in my opinion, most important considerations is the actual installation experience. Simply getting the system up and running out in the field is often more complicated in practice than it seems on paper. Your goal should be to make the field installation as simple and smooth as possible. You can’t take a simple process, make it more complicated and error prone and expect to see widespread adoption. Let’s say the existing process is to mount the product and connect three wires. If that process takes the installer 30 minutes, that’s your baseline. The “new and improved connected product” will likely involve some workflow changes, but you should be ultra-sensitive to how you impact the current process. If that same installer now needs to bring a laptop and a USB cable and find a 25-character Wi-Fi password which must be typed in by hand … you see where I’m going. You’ve introduced additional variables into the process that are prone to fail. At some point, the laptop battery is going to die or the Wi-Fi password changes and this installer’s productivity level tanks. You end up spending all of your time on the phone providing support rather than creating the next great feature.
Each step of the process will also include some sort of B2B data integration. Whether you’re procuring parts, loading firmware, matching device identifiers to shipping records or activating device connectivity, you’ll need to communicate with a backend process. Enterprise integration for supporting systems monitoring inventory, scheduling service visits and billing will likely make up a significant portion of the overall professional services required when commercializing IoT products.
Consider the task list required to meet the analyst predictions for 2020. If you expect, on the low side, 25 billion connected things by 2020, we have an amazing amount of work to do. I’m talking about all of these logistics tasks required to simply deploy billions of things. Even accounting for the 4.5 billion already connected, if we started immediately, we’d have to connect a little over 11 million things every day for the next five years to meet those expectations. The logistics alone is a big mountain to move.
I believe the mountain is already moving, but as more people jump at IoT opportunities, it’s important to remind the community that we’re not shipping ones and zeros across a wire. IoT is about shipping physical products that require the coordination of people across several organizations and disciplines. It’s imperative that we consider commercializing IoT and the workflow beyond the basic technical use case and realize that small decisions made in one area can have a huge impact on the process elsewhere. Keeping that in mind will help ensure everyone who has to interact with the product — across the entire spectrum — has a good experience.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.