Enterprises commit these three costly oversights when launching smart consumer devices

Over the last couple of years, the pace of enterprises looking to enter the smart and connected device market has accelerated. Ranging verticals from HVAC to home appliances to security systems, OEMs are building and launching IoT-enabled versions of their unconnected counterparts. While selling and marketing to consumers is not new to these companies, meeting consumer expectations of smart-home enabled devices in a crowded space presents a set of new and unfamiliar challenges.

One of the areas where enterprises tend to fail to adequately develop is the cloud platform, or the backend, that enables and supports the fleet of connected products. The cloud component of an IoT solution tends to be under-appreciated with the spotlight often shining on the more tangible and visible components such as the mobile app and the device itself. This is especially when everything is working as expected. However, as enterprises bring their IoT devices to market, and begin to scale, it is often the cloud that makes or breaks the solution.

In this post, I’ll touch on three often overlooked aspects of the cloud and explain why they are important to the success of a consumer IoT product.

OTA Update Management

OEMs that bring connected products to market need to be ready to update those products periodically and do so at scale. While it varies by product category, consumers tend to expect that smart products get smarter over time. In addition, connected products may need to be updated by the device maker to fix bugs, to patch newly-discovered vulnerabilities, to improve stability and performance, or to reduce cloud costs.

Even before a product hits the shelves, there is oftentimes an update already awaiting. As devices come online, enterprises need a systematic way to update those devices to the latest and greatest. Needless to say, given consumer IoT scale, this process must be run without a human in the middle since users can choose to unbox and connect their purchase at any time.

Then, once a new update is available, the enterprise must systematically deliver this update to devices in the wild. Generally, this is done via a staged rollout that starts with a small cohort, and then if everything is progressing to plan, continues with larger groups of devices until all devices have received the update. That is until all connected devices have received an update. The OEM needs to also ensure that any devices that are disconnected, or are otherwise unable to take an OTA update, eventually get the update at the next opportune time.

And given that a 100% success rate is difficult to achieve, it’s important to be able to remotely diagnose any OTA failures and take the necessary steps before a retry.

Remote Troubleshooting

Companies launching smart products need to be prepared to solve all sorts of issues remotely, often during a live interaction with the customer. Whether it’s with the actual product or anything else that might be getting in the way, such as home router configuration or interactions with smart home ecosystems like Google Home or Amazon’s Alexa, consumers expect to get support.

While pre-launch devices are being developed or tested, it’s easy to find and fix issues. In those situations, engineering often has perfect, or near-perfect information about the environment, and can ask user-specific questions, as well as take an hour, a day, or even multiple days to figure out the root cause of the problem. To make matters worse, any dogfooders that are too close to product development may know ‘too much’ and may not be representative of an archetypal user and may not run into issues that will be seen in the field.

The ability to help a customer troubleshoot, as-if looking over their customer’s shoulder, relies on a set of cloud features that often doesn’t get enough attention while the product is in development. Once products start to ship, customer service lines begin to light up with routine issues as well as unforeseen problems that need to be resolved to prevent exciting customers from turning into frustrated ones.

Cloud Cost Model

Enterprises looking to get into the consumer space should have a refined cost model that projects the cloud cost of a device in the field.

Given the many technical challenges involved in getting an IoT product to market, an often forgotten element is managing the ongoing operating cost of in-market products. This ranges from the cost of cloud services to the cost of human capital needing to maintain and secure and performant cloud platform, as well as the cost of developing new cloud components such as those discussed above. This often results in sticker shock to product and engineering teams that previously haven’t had to budget for operational costs and continuous capital investment required to support IoT devices. This is especially true for data-hungry devices such as video-consuming security cameras and baby monitors.

While many consumer-grade devices tend to go obsolete rather quickly, there is no easier way to draw the ire of customers than to try to sunset cloud-dependent products while they are still in use.

Summary

Enterprises looking to launch successful smart consumer products will need to develop a means to manage OTA updates at scale, enable remote troubleshooting of devices in the field, and ensure their connected solution considers the holistic economics of selling and supporting a smart product.

In future posts, I’ll dive deeper into these components, and share other insights into the IoT market.