One of today’s hottest technical trends is the Internet of Things (IoT). The idea is to network together physical objects containing electronics, software, and sensors on the Internet so they can collect and exchange data, offering improved efficiency, accuracy, and economic benefit. Each “thing” is uniquely identifiable through its embedded computing system (that is, by its MAC address), and can interoperate within the existing Internet infrastructure. Experts predict nearly 50 billion objects will be online within five years. The possibilities seem endless (here is a good rundown) and, frankly, a little bewildering.
My former colleague Wynn Grubbs, now senior vice president of business and partner development at PlumChoice, recently posted an article reporting that consumers “struggle to use common connected devices.” Talk about bewildering! Fully two thirds of consumers are “confounded” trying to install or use their purchases. When they need help, half the time consumers only ask friends and family; of those who turn to technical support, 80% are dissatisfied with the experience. Nor is the problem confined to older consumers: 60% of Millennials, the vaunted generation that literally grew up with computers, struggle with their own smart devices. (There is no mention of anyone turning to documentation.) One seventh of consumers eventually return their purchases.
That return rate (14%) is almost triple the return rate of consumer electronics in general, which is already too high. As I previously discussed, Accenture found that customers return 5% of consumer electronic devices because they “don’t work,” even though 95% of the devices returned actually work perfectly. Support, replacement, and restocking costs all reduce the profit margin, not to mention the cost of dissatisfied customers, who rarely come back. Returning items that work correctly is an expensive irony.
Now, the first generation of programmable devices featured unique hardware interfaces, and consumer acceptance was hit or miss. For example, I figured out how to use the brass springs and stops on my lawn sprinklers to water my grass but not my windows, but I know people who just let theirs spin. Smart devices are obviously much more complex and sophisticated than their “dumb” ancestors, with more functions but also more failure modes. The manufacturers’ instincts are to cram as many “smart” behaviors in as possible to justify the cost. A “smart” lawn sprinkler might be able to sense the actual rainfall and the state of dryness of your lawn, saving water and money by watering only when your lawn needs it (and sparing you the effort of going out and turning it on). Perhaps it can avoid watering at night, get the weather forecast from the Web, or use your community’s website to avoid watering during mandatory conservation days. There is probably an app that goes with it, so you can water the lawn from wherever you are. And perhaps it even has voice recognition, so if you say “the grass looks dead” (or “get off my lawn!”) it turns on. But such a device might fail to work altogether if the moisture sensor fails, if it can’t maintain a Wi-Fi connection, if the battery runs out, if a software update fails to install correctly, if the sprinkler’s CPU crashes, if the app fails, if you forget your account password, if it can’t make out what you’re saying, or even if the system is, however improbably, hacked. (Not to mention that running over it with the lawn mower just got a lot more expensive…) In the Internet of Things, where all devices are connected, a brown lawn might require you to buy a better router.
To get consumers to understand how to make sense of their smart devices, manufacturers have tried to make them as simple to use and autonomous as possible. However, this has only hidden their complexity and the potential problems behind a simplified GUI.
Now, understand that Wynn is offering his company’s support services to manufacturers. I wish him good hunting. But we too play the game of helping people successfully understand and use technical devices. (That’s how we make the world a better place.) What can we do? I see both front-end and back-end opportunities in the IoT.
Since the PlumChoice press release mentioned smart thermostats, and since I started my career at a company famous for its thermostats, I’ll take as a case study the Nest Learning Thermostat. (I don’t own a Nest and I’m not suggesting the product has any particular issues; in fact, Nest customer ratings on Amazon are very high.)
Back in the day, the “dumb” thermostat had one sensor and did one thing: turn on the furnace when the room temperature dropped below a set level. Then it learned to do two things: turn on the furnace when the house got cold and the air conditioning when it got hot. The second generation of thermostats was programmable, so that the temperature could be changed at specific times during the day and week: cooler at night to save money in winter, warmer during weekdays in summer while everyone was out. Home heating and cooling is a major component of home costs and energy usage, so if you’re willing and able to program one, a $50 thermostat will still save you money and energy and pay for itself very quickly. Why spring for a connected thermostat, then? The Nest business model is that consumers often can’t figure out, or can’t be bothered to closely program, their programmable thermostats. They’re right. As How-To Geek says in its product review:
You know what we hated the most about our old programmable thermostat? Even if you had memorized the arcane and numerous button combinations required to program the device it still took a significant amount of time to reprogram it which meant you were left standing there in the living room, your arms losing feeling, poking away at it for 15 minutes or longer anytime you wanted to do any significant reprogramming.
Even so, for five times the price, the Nest had better do more than a programmable thermostat. It does. The Nest includes a motion sensor to detect when you’re in the house, and automatically compiles a schedule. It remembers your temperature overrides to learn what temperature you find comfortable. It knows how long it takes your heating and A/C systems to work and takes into account the lag time.
How do consumers learn to use their Nest? From what I can see, the company provides a brief installation guide, important for a product that must be hard-wired into a home electrical system (some consumers pay for certified third-party installation), and only a ten-page “welcome guide” (five pages in English, five in Spanish) that is no more than a marketing pamphlet. The primary information vehicle is the Nest website—and YouTube. For example, here’s a setup video that as of this writing has been viewed 360,000 times on YouTube; I should have such numbers for my work! While I would be comfortable communicating the information in a technical document, the video is more appealing and, I judge, more effective.
Consumers don’t instinctively understand how to use their IoT devices, so the need for technical information still exists, but the delivery media have changed. If your skills run to website content and video, the future look promising. But if your skill set is writing technical manuals, you won’t have much luck working for companies making IoT devices—at least, not on the customer-facing side. Fortunately, there’s someplace else to look.
These sophisticated devices have a back-end Internet interface. Consumers aren’t expected to configure network connections themselves; the devices must make reliable and secure Internet connections on their own. Chris Ciufo neatly summarized the technology and the challenge behind designing smart-device connectivity:
Most of these embedded “wannabe nodes” were created by engineers who’ve never before designed with Wi-Fi. Nor do they understand the hundreds of APIs needed for the most basic TCP/IP connection.
Or: how likely is it that designers have experience with IoT security requiring lock down to protect factory automation or your nanny cam? Forget it; Wi-Fi’s 3AES and the Internet’s TLS/SSL security is more complicated than the whole device itself!
Chris is saying that even the design engineers need help understanding how to get the connections to work. Personally, I like the sound of “hundreds of APIs,” and I think technical documentation has a role to play behind the scenes as well.
So, then, there is a growing market for IoT devices; consumers have ongoing difficulties figuring out how to use them; manufacturers want an inexpensive and effective way to communicate technical information to both consumers and designers; and here we are. I think this makes sense.