Why is my smart device making me feel dumb? The challenges of onboarding IoT devices

James wants to water his lawn. He picks up the hose, turns on the water, and a few minutes later…the job is done.

Then James installed a smart sprinkler system.

Before his lawn can be watered, James needs to make sure his WiFi is enabled, his device is configured to his municipality’s water use guidelines, the programming is set properly to water Monday, Wednesday and Friday from May through September, and the device is connected correctly to other information systems like The Weather Channel so the sprinkler will not be activated if rain is in the forecast.

Whew. Maybe he should have stuck with the hose.

The Consumer Challenge

While in theory, smart devices should be easier to set up — most are wireless and can be remotely connected; in reality, the setup of smart devices can present significant challenges to the consumer. Successful IoT onboarding is dependent on the consumer to choose the right location for the device, to connect correctly to the current infrastructure and the Internet, and to configure the device properly to the associated app(s) and the overall ecosystem.

Early adopters of smart home products tended to be DIY-ers or highly technical and were able to self-onboard with ease. However, recently smart home adoption has been slower than anticipated. In order to more efficiently shift the usage of smart devices into the mainstream population, the IoT onboarding process must be simple, seamless, and highly personalized.

As the IoT matures, and the range and complexity of connected devices multiplies, there is a need for new approaches to support IoT device onboarding. According to a recent survey by TechSee, 76% of customers prefer self-installation for home electronic devices, and visual guidance has emerged as the solution of choice for the majority of customers.

The Enterprise Challenge

Companies have taken notice of consumer preference for self-service. Brands are reevaluating the current Support of Things (SoT) customer support models in light of this inevitable shift. More and more enterprises recognize the ROI potential from significant savings on labor costs, reduced technician visits, lower call center volume, as well as the ability to scale SoT services.

After all, with the number of connected devices expected to reach 20.4 billion by 2020, and their relatively low cost / profit, companies must find a way to better balance their customers’ need for assistance with installation, activation and troubleshooting. Clearly, dispatching a technician at a cost of $150 to install a smart lamp that costs $30 is not feasible.

Enter AI-powered self-service solutions for smart home IoT onboarding.

Traditional self service

Today the most common method for onboarding smart home products is via the use of self service tools, and especially how-to videos. YouTube reports an increase in product unboxing video views by 57% in one year, and an increase in uploads by more than 50%.

It would take a viewer more than seven years to watch the huge number of videos that have been uploaded to YouTube in one year alone, and these videos have more than a billion views annually. Google Consumer Survey underscores these statistics, with 20% of consumers (1 in 5) reporting that they’ve watched an unboxing video.

However, despite the popularity of videos, there are still gaps. The process that seemed so simple online can be much more difficult to execute at home. If the customer is technologically-challenged, or the model is slightly different, or the environment does not match the one in the video, the customer experience can be negatively affected. The fact is, despite of the growing availability of unboxing videos, customers still need to call the contact center for assistance with device installation and operation.

Auto onboarding

A relatively a new solution, auto onboarding is executed when the customer scans the QR code and connects the device to the Internet, enabling the set up and activation of the product to be completed automatically. For example, Intel’s Secure Device Onboard service uses a “zero touch” model that allows devices to dynamically discover the customer’s IoT platform account at power-on for automatic registration.

While this may be an ideal solution when it comes to software and connectivity aspects, auto onboarding does not deal with the hardware aspect — cables, placement, and hardware configuration. And it is these components that consumers find the most challenging according to TechSee’s survey: cabling/wiring (28%), hardware configuration (16%), and component identification (12%).

Virtual Assistants

Conversational AI platforms — known as chatbots — automate and scale one-on-one conversations, and are increasingly becoming more sophisticated. As they grow smarter, IoT chatbots can be more helpful at diagnosing service problems or walking customers through activation steps, for example. The popularity of voice assistants is also on the rise. A full third of consumers already prefer self-installation help via interactive conversational virtual assistance through tools such as Alexa, Siri, and chatbots. Alexa has already proven capable of controlling smart home cameras, entertainment devices, lights, locks, and thermostats, and others.

Both tools are effective in interacting with consumers using the ability to process text or voice, and retrieving related information from a knowledge base. However, both of these assistants still lack the ability to provide feedback to the customer or correct him when something goes wrong, and sometimes fail to deliver user experiences that are seamless and efficient.

Visual Assistants

Visual assistants can take IoT device onboarding to the next level — they can bridge the physical and digital by allowing the computer to “see” the customer’s environment. Using computer vision, visual assistants can process images of the customer’s home and understand the status of the customer’s onboarding issue: whether it’s installation or troubleshooting. This capability allows visual assistants to fully interact, provide feedback and correct the customer when necessary.

For example, when unboxing a new smart security system, a visual assistant can recognize the cables and inputs, and guide the customer using AR as to which cable to put where. The bot can see the customer’s physical space, advise the optimal locations to place the cameras, recognize LED lights of the cameras, and help the customer complete the installation.

Summary

To boost the smart home revolution, IoT device onboarding should be simple and seamless for the consumer, and cost-effective for the enterprise. The need for advanced AI-driven self-service onboarding solutions — such as traditional self-service, auto onboarding, virtual assistants and visual assistants — is evident and expected to increase exponentially as more and more consumers bring a wider range of complex connected devices into their homes and their lives.

This post was initially published on the TechSee Blog