Challenges in designing a connected product

The Internet of Things (IoT) is changing the way companies build and market products, and it all starts with product design. As an engineer, the most important thing to keep in mind is that you're no longer merely designing a product; you're designing a system with multiple component parts. Those components might include the product itself, a mobile app, cloud-based software, a router, an IoT platform, and more. Connected products are inherently more complex and require a new focus on system integration.

To successfully design a connected product, big-picture thinking is required as well as a focus on crucial details. End-to-end testing is a must to ensure that all components work together seamlessly, and the leadership and organizational culture must be in place to support extensive testing and ensure success. The initial failure of the Hubble Telescope is a good example of what can happen when these elements aren't all in alignment: After a 15-year, $3 billion project to assemble the most advanced telescope in history, Hubble drastically underperformed due to a flawed mirror, which wasn't caught before launch due to inadequate testing.

NASA was able to put together a solution, and the Hubble has been expanding human knowledge for 25 years now, but the initial failure is a case study in how projects can go awry due to an inadequate focus on testing. Another key factor to keep in mind during the engineering process is that the systems involved will keep evolving over time. With a connected product, the engineering team has to account for the evolution of platforms (e.g., iOS and Android), routers and other components, all of which have to be a part of the testing regimen.

It's also critical to understand that companies that create connected products are undergoing a transformational shift in their business model that will affect product design. The IoT connection opens up an ongoing relationship with the customer—it's no longer about driving a one-off purchasing transaction. Now the company can send updates to improve product functioning, expand the options available to customers, and create partnerships with other companies or complementary product lines that work with the original product.

This transformation requires new ways of thinking about the product lifecycle, and the customer support infrastructure will have to be adapted to accommodate the connected product. The pricing model may also require updating. Product manufacturers will have to gauge the appetite for subscription fees, determine a strategy to support hosting costs, and decide whether it makes sense to partner with a cloud-based IoT platform provider or build their own solution. That decision can have major implications since an in-house approach requires a fundamental expansion of the company's core areas of expertise.

When designing connected products, engineering teams have to think in terms of a distributed system and identify system partitioning needs for the product, cloud, and app. They'll need to make decisions about where functions reside and agree on an optimal user interface strategy. Should buttons go on the device or in the app? Should the device user interface be extremely simple—just an on/off button—with additional functionality located in the app? These questions and more must be resolved. One factor that should be considered is the possibility of pushing out updates and bug fixes through the cloud. It's a future-proofing strategy that can improve product "stickiness," reduce costs, and cut liability exposure.