I got my smart TV in early November 2016. It was a 50-inch Samsung—the largest size I could convince my then skeptical roommate to shell out for. We scored a modest deal. At roughly $600 split between two poor millennials, it wasn’t half-bad. (Though, in retrospect, I should have waited until Black Friday.) It was, and I mean this in the most generous of terms, okay. The picture quality was alright. The Netflix, Hulu, Amazon Prime, and HBO Now apps all worked. Screencasting was sometimes choppy but altogether fine. And then last year, I moved.
I was a slightly less poor millennial but in the interest of saving a couple of hundred bucks, my partner and I schlepped this three-year-old TV in the back of a Prius from Brooklyn to Manhattan. It was fine. And then suddenly, it wasn’t. The apps, which rarely had issues before, started crashing every few minutes during whatever episode of TV I was watching. Streams started buffering at painfully slow speeds. At first, we wondered if it was our internet—but a secondary TV purchased around the same time had no issues, and we’d invested in a faster internet service and a particularly beefy router. It didn’t matter how many times I disconnected and reconnected my wifi. I downloaded all the updates. Plugged the TV in and out. It wasn’t broken, per se, but over the next few months, my trusty old TV had morphed into a glitchy, nigh unusable pain in the butt. By the time Black Friday 2019 rolled around, we’d had enough and bought a new one. I was annoyed but unsurprised. I’d seen this play out with many smart gadgets—to the point where it sometimes feels like I’m renting tech and periodically paying upgrade fees, not truly owning it.
While this was all happening, I thought back to the blocky CRT television my father had kept in the living room for my entire childhood and most of my teen years. He didn’t upgrade until he too, moved after I’d graduated from college. That thing had lasted him at least 15 years. What happened?
This is a familiar story, especially if you’ve purchased a smart TV in the past few years. You plop a decent chunk of cash down for a gadget, it works so much better than your old one for about a year or two, and then slowly but surely it starts to suck. The battery doesn’t last as long. Loading screens or apps gets laggy. Glitches happen more frequently. At a certain point, you sit there, scratching your head, your brain doing mental calculations of whether your bank account can handle an upgrade. Sure, I’ll fork over $1,000 for a smartphone every three years because planned obsolescence with phones is real. There’s no way in hell that I’d ever shell out $2,000 for a Samsung smart washing machine if there’s even the slightest chance I’ll have to replace that thing in under ten years.
This is the conundrum with the Internet of Things. You, the consumer, only buy appliances when they crap out on you. If you shell out for an advanced gadget, the last thing you want is for it to break like a cheap piece of trash. You also don’t want a device that’s going to promise the moon, only to get discontinued because the company went out of business or the product line wasn’t profitable enough.
The most recent public example of this is Sonos. Back in January, the company announced that its oldest products would be effectively retired and rebranded as “legacy products.” The backlash was fierce and immediate. It didn’t help that Sonos said 92 percent of all the gadgets it had ever shipped were still in use. While the company probably meant that statistic to emphasize it had put effort into building quality products, some users claimed the move was a cynical cash grab. Users had simply held onto their devices too long, and Sonos stood accused of sunsetting, or intentionally phasing out, perfectly usable products to drive up profit margins. To be fair, ten years is a long time for any consumer electronic and as Sonos pointed out, the limitations of 10-year-old processors as newer streaming technologies emerge are very real. But as many audiophiles will tell you, a good pair of speakers can sometimes last you twenty years with regular maintenance. To many consumers, speakers, like appliances, are a long-term investment.
There was another part to the Sonos saga, however, that illustrates the problem with the future of IoT as a whole. Updates. Sonos told customers they had the option of keeping their legacy products—so long as they acknowledged they wouldn’t get updates or security support after May 2020. Critics rightfully pointed out this was a major security hazard. Moreover, users also questioned how newer and legacy Sonos devices would even interact past May. For many, the delineation would break their carefully crafted setups for “no reason.”
Sonos backpedaled somewhat. Their older products would still be phased out, but would receive bug fixes and security patches “for as long as possible.” In a blog, Sonos CEO Patrick Spence personally apologized for the flare-up, telling customers that the company had heard their complaints and admitted they had “not gotten this right from the start.”
The brouhaha wrapped up much like how you’d expect it to. After Spence’s note, Sonos most recently announced it would be releasing a new app in June. Sonos products launched after that date would exclusively work on the new app, while legacy devices will continue to work on the original app. Non-legacy devices released before June would be compatible with either the old or new app. However, users will eventually have to choose which app—the old or new—they’re going to stick with going forward. Sonos presented a number of options, but the simple truth is there’s no elegant solution here.
What makes the Sonos case interesting is this was a recognizable, reputable brand and regardless of how you feel the company handled the matter, it made what appears to be a good-faith attempt to communicate its dilemma with its user base. Gizmodo reached out to Sonos to see if the company would comment further, but the company declined, stating it had nothing further to say.
Sonos’ dilemma of how to update products, stay transparent, and somehow still turn a profit isn’t an isolated problem in the IoT space. It’s just happened to be the most visible and illustrative example in recent memory.
Case in point, Philips Hue’s smart light bulbs are the gold standard when it comes to connected lighting. The individual bulbs ain’t cheap, but anyone who shells out for a Philips setup will tell you they work well, integrate with plenty of third-party apps, and are convenient as hell. But since launching, Philips has upgraded the wifi bridge needed to group these lights and enable many of its more advanced features. The company will end support for the first generation bridge at the end of April this year.
That means if you haven’t bought the second-gen bridge, you’ll need to soon. That’ll only set you back $60—an annoying fee but not one that’s likely to spark outrage, like a $300 to $500 speaker might. The gap between the first and second bridges was only three years. By the time support for the first bridge ends, it will have been around for only five years. Philips’ website says that bridges will get at least three years of updates and supports after being replaced; the lightbulbs that are meant to last 25 years and cost anywhere from $15 to $50 apiece will get at least 5 years.
At the very least, a regular LED lightbulb doesn’t pose a security risk if the company that made it runs out of money. That said, it’s likely there will be a third-generation bridge at some point down the line—and at that point, you’ll have to wait to see if all your old bulbs are still compatible and then calculate the cost of upgrading the bridge, replacing non-compatible bulbs, and consider if that number is worth the effort. Calculating whether it’s worth sticking with an IoT ecosystem is annoying, even when the stakes are ‘small’ like with Philips Hue.
The uncertainty around an IoT product’s lifespan (and its afterlife) is a problem that grows exponentially the more expensive the gadget. Lightbulbs are relatively cheap in the IoT space—you could bear the brunt of the maintenance cost and not go bankrupt. The same can’t be said for more expensive gadgets that are starting to find their way into homes. Robot vacuums with wifi connectivity and advanced navigational tech can cost a pretty penny at $700 to $800 on the high end, and for that much, you want that thing to last more than two years before it craps out or is discontinued. Dyson’s 360 Eye robot vacuum cost $1,000 when it launched in 2015, but by 2018, you could only find it on Best Buy or eBay. Many companies continue to support IoT gadgets after they die, but not all do. While Philips and Sonos both provide a guarantee of how long they’ll continue to support a dead device, it’s not an industry-wide standard to do so yet. Many die a slow death marked by increasingly infrequent updates and sputtering functionality. The best-case scenario is you’re left with a dumb device you paid a premium for.
Right now, this isn’t a widespread problem. Most IoT appliances are still stupid expensive for the average person. But if Sonos is any indication, this is a problem manufacturers ought to solve sooner rather than later. The IoT market was estimated at around $212 billion by the end of 2019, and is expected to grow to $1.6 trillion by 2025. And, every year Amazon’s list of connected hardware for the home only grows bigger and more affordable. Sure, the IoT is still a nascent industry. You could make the argument that by the time we can all afford smart fridges, washers, dryers, and cars, this will all be moot. Maybe this is a problem for the dozens of small companies with dubious funding.
“Ultimately, honestly, it’s going to impact the big companies too,” Frank Gillett, an IoT analyst at Forrester, told Gizmodo over the phone. “Netflix is pulling some of its apps off some older Roku devices. You buy the Roku expecting that Netflix will always be there, and then Netflix says we don’t feel like keeping our code state compatible with old Roku devices.”
“Part of this is an expense thing,” Gillett explained. “It becomes more expensive and more complicated for a company to continue supporting old technology. This was the best they could imagine 10 or 15 years ago. And frankly, part of the challenge is the customer last paid them 10 or 15 years ago.”
For Gillett the one major challenge with IoT products is the shift from a one-time transactional relationship with customers, to a continuous one. In ye olde days, you bought a microwave and that’s the last you thought of that company unless there was a recall. There was a clearer relationship between the cost to make and design a product, and the profit derived from the sale. Now, as more devices become connected, companies have to shift their mindset to that of a service company. Long after you’ve paid for a device, the company must now also pay the cost of maintaining an app and servers. Suddenly, the more devices that are sold, the more money a company must spend—all while not bringing in any more new revenue.
In other words, with connected gadgets, it doesn’t pay for them to last a super long time.
The IoT is still a nascent industry and clearly, there’s a disconnect between customer expectations and the realities of capitalism. It’s also distinctly possible that as connected gadgets become the norm, owning things will become a relic of the past.
“It’s always going to be more expensive to build a product, so the basic conundrum that I described will persist,” says Gillett. “But what I think will happen eventually is the market will solve the problem in a different way, because I’m not sure consumers are ever going to wrap their head around the idea that they need to subscribe to their dishwasher.”
The example he gave me was in the kitchen. Nowadays, even meals and groceries are viewed as a service—think Blue Apron, FreshDirect, etc. It’s not a stretch to think that maybe a grocery or food company will begin to offer some kind of subscription model that also involves hardware. Maybe you pre-commit $300 a month to Wegmans, for example, and that buys you a higher level of service and convenience. You get so-and-so number of prepared meals and customizable grocery delivery options. Eventually a smart appliance—say a fridge that can monitor when you’re running low on items—becomes part of that. Suddenly, you don’t have to worry about shelling out or maintaining the appliance—just whether you can manage the additional monthly fees.
You can already see this happening with services like ZipCar. Owning a car is expensive thanks to insurance, loan payments, gas, and maintenance costs. With something like ZipCar, you can subscribe to car ownership without having to actually own a car. Uber and Lyft have also gotten in on the subscription game.
There are other early examples of service-gadget bundling. Nomiku, for one, was a startup that helped popularize affordable sous vide immersion cookers. One product, the Nomiku Sous Chef, was an immersion cooker that cost a relatively affordable $150. The catch was, to buy it, you also had to subscribe to Nomiku’s meal plan. Entrees cost anywhere from $8 to $14, and side dishes ranged from $4 to $6. You could mix and match meals, customize dietary preferences, and there was an RFID element that acted as an automated inventory tracker. After you bought $300 in food, the company would credit you $150—or basically, the cost of the device. The company shuttered in late 2019, mainly because of an overabundance of affordable sous vide options—and, crucially, the meal plan part faced logistical challenges. It never launched outside California.
Another example is the Tovala Oven, which is still around. Basically, it’s a smart oven that can automatically scan brand-name items for the perfect cook time. So, you get a DiGiorno’s frozen pizza from your grocery store, scan it, and wham bam you get a pizza. But also, the company sells its own meals—and if you sign up for a certain number of meals, the oven’s $300 price tag gets knocked down $100. Only time will tell if this goes the way of Nomiku—smart ovens haven’t quite taken off just yet, though there are quite a few out there.
Still, it’s easy to look at both Nomiku and Tovala Oven and think, hmm...what if Amazon copied it, knocked a few hundred dollars off the price, and bundled it with its Amazon Fresh service? Suddenly, it seems less fantastical. Dystopian yes, but also awfully convenient.
That future isn’t quite yet set in stone—though it is one that companies are currently experimenting with to varying degrees of success. In the meantime, the IoT has plenty of other problems to solve. Mainly, in the form of security and what happens when a smart home product is either dead or fading into obsolescence.
Recently, the UK introduced new legislation to strengthen IoT security. It outlined three rules that all manufacturers would have to abide by: 1) All products would need unique device passwords; 2) Companies would have to provide a point of contact for reporting bugs; and 3) An explicitly stated minimum length of time that products will receive updates at the point of sale. The U.S., meanwhile, has made some attempts at legislation but they’re overly vague and only address connected devices used by the government.
The UK’s proposed rules, while reasonable, still don’t address the issue of what happens when the connected item you bought is discontinued, the company goes out of business or is hacked, and no one else is out there willing to take over the business. And at the end of the day, that’s the problem with smart home gadgets. Some, like a robot vacuum, will technically function in some capacity if you stripped away all connectivity. In Sonos’ case, you still get legacy devices that work with a legacy app for an undefined duration. Others, simply won’t. A smart alarm system, for example, is just bits of plastic and wire if the servers go down. In either case, you’re left wondering if you’ve been ripped off. Until manufacturers start guaranteeing a minimum level of offline functionality, the smart home feels like a neverending exercise in cost-benefit analysis.
Funnily enough, our secondary smart TV—the one that was fine when our old one started breaking down—has also started dying. The Hulu, Netflix, and Amazon Prime apps crash at least four or five times during a half-hour of television. Whenever I turn it on, it screams in static; a sudden development that I have no explanation for. It keeps dropping wifi signal despite being very close to the router. We’ll probably end up sticking it out until the next big shopping holiday, but I am already wondering if there’s a point.
Correction, 03/24/2020, 1:40pm: An earlier version of this article spelled Frank Gillett’s name incorrectly and has been fixed. We regret the error.