CES 2013: Connected Devices and the Internet of Things
Photograph by Jacob Kepler/Bloomberg
The Internet of Things and connected devices will be all the rage at the Consumer Electronics Show next week in Las Vegas, as companies seek to wow the audience with connected fridges, iOS-based home-automation products, and dongles galore that will connect everything in your home to some wireless standard or another. But amid the excitement and heady rush, it’s worth understanding what the Internet of Things really is and how it will make money.
To that end, here are a few handy questions about this developing technology: These aren’t exhaustive, but they are a start, and I find them especially useful as we see more and more projects on Kickstarter, hyped stories about cool devices, and venture capital deals galore. Feel free to share your own questions or thoughts as well.
Connected devices are nothing without a service, so where is the service? There are plenty of connected devices out in the market, from Fitbits (see disclosure below) to garden sensors that use a home’s Wi-Fi network to tell you when to water your plants. But the key to building out a true Internet of Things experience isn’t in being able to connect devices to the Web, but building a service based around that connectivity.
A connected fridge is worthless if it can’t help manage your grocery list. A connected pedometer that doesn’t link calories burned to a service that also tracks calories consumed isn’t much good. And those services are where one can find more revenue. For example, if you can match a connected pedometer and calorie-tracking product with a service that plans meals based on a person’s activity, that’s something many people would pay for. They might pay even more if you deliver those meals precooked or as a cluster of premeasured ingredients bought via subscription.
How many Internets of Things will we have? Many people think of the Internet of Things as some magic web of connected devices that will communicate with each other and act together, but the reality is probably closer to the vertical segmentation we already have in our lives. So while we might have an Internet of electricity that combines elements of the smart grid with our thermostats, we may have to buy some kind of device that we plug our appliances into to connect them to the Internet of electricity. Likewise, our cars might be connected to our home-entertainment networks via Wi-Fi or a smartphone, but if we have self-driving cars, they will also need to connect to an Internet of vehicles or transportation that will include other cars and traffic signals.
Where will the intelligence in the Internet of Things live? Most connected devices are part of a limited ecosystem with only one or two devices, but as the sector matures, will they communicate all the way back to a server in the cloud? Or will there be some kind of in-home or in-car (or in-whatever) device that helps make decisions based on the information the connected devices deliver? If you lose your connection to the Internet, having on-site intelligence keeps your home or car network operational or at least communicating with each other as locally networked devices. It would suck if your door locks didn’t open when your broadband was out. The counter to this argument is that many of the in-environment devices are proprietary, so it’s impossible—or at least expensive—to connect them to everything.
Where are the toll bridges in each ecosystem? Once you view the Internet of Things as many Internets and think about standards for connectivity, software, and even the clouds where data might be stored or processed, it becomes clear that each ecosystem will have different control points where it can make a profit. Sure there’s money in selling the physical hardware if you’re Apple (AAPL), or maybe Nest Labs, but there are also services, data, and online processing or storage that could also provide revenue opportunities for companies playing in the Internet of Things. I could write an entire story about possible business models, but, just to get you thinking, ponder the example of getting your fridge to buy your groceries.
In that use case, to build the service, the maker of the fridge or a third-party entity will have to make deals with grocery stores for access to their pricing and inventory APIs. That API would be a potential toll point. But does the grocery store have that API, or is there an opportunity for other companies to build it? Or even aggregate such data from a variety of stores and then offer it to those wanting to build such a service? Some startups, including Grocery Server, which has created an API that taps into store merchandising systems, are building up such databases of pricing information. The question then becomes: Does the builder of a service want to work with one provider, or go direct to the source and work with many?
Of course, if the connected device itself is popular (like an iPhone or an Xbox), you can dictate the terms and means through which people could interact with your product. In the case of the fridge, all of the appliance makers might come up with a standard for pricing and how grocery stores label their food, thus totally controlling the development of the entire ecosystem.
That’s probably enough examples—and enough questions—for the time being. The point here is that while people spout a lot of information about the Internet of Things, as a business opportunity there are still a lot of unanswered questions. And until we start asking those questions and getting answers, it’s hard to see how useful some of these cool projects on Kickstarter will be, or who will win in developing ecosystems.
Disclosure: Fitbit is backed by True Ventures, a venture capital firm that is an investor in Giga Omni Media, the parent company of this blog. Om Malik, founder of Giga Omni Media, is also a venture partner at True.
Also from GigaOM:
CES 2012: A Recap and Analysis (subscription required)