Sonos should take the ecosystem approach to old devices

Sonos, the maker of internet-connected speakers, experienced a PR disaster this week caused by its failure to recognize how connected devices change the hardware landscape. Sonos isn’t the first company to experience a backlash tied to changes in how it treats older connected devices, nor is it the first to try to walk those changes back.

But it is one of the largest companies to experience this particular nightmare, and its decisions should influence the conversation around regulation and adoption of smart home devices for years to come.

On Tuesday, Sonos dropped a bomb with an email letting owners of older gear know that their products would no longer receive software updates. The email laid out the options for those affected, namely that those owners could exchange their so-called “legacy” gear for new products at a 30% discount, or they could leave it as-is.

However, for those opting to do nothing, Sonos writes: “Over time this is likely to disrupt access to services and overall functionality.” The email also added that people who kept the legacy devices mixed in with newer devices on the same network would lose upgrades to the newer devices.

Sonos customers took their rage to social media to admonish the company. As someone who has spent $2,000 on Sonos gear since 2009, I was among those feeling hard done by. Sonos stock fell 4% in the 24 hours following the announcement, which is fairly unusual. Generally, Wall Street doesn’t mind a business-saving move that pisses off customers.

On Thursday afternoon, Sonos sent a more placatory email in which it tried to stress that legacy devices would continue to work and that it would continue to provide security updates. It also made a new promise, writing: “If we run into something core to the experience that can’t be addressed, we’ll work to offer an alternative solution and let you know about any changes you’ll see in your experience.”

Notice the email doesn’t and — truly can’t — say that if something breaks in the future, Sonos will fix it. Rather, the company will try to offer a solution and let the customer know if something will change. But while that’s a start, it’s not enough. Consumers and regulators should take the experience of Sonos and other recent device shutdowns and turn them into a rallying cry for products to be remade for the connected era.

The Sonos drama has led to headlines about the end of ownership and the failure of the smart home. Neither of which is a bad take; I’ve written similar stories. But what we have to accept (or maybe we don’t, but I’ll get to that later), is that the products we buy will increasingly have some sort of internet connection and smarts. And when you put smarts into products, you shorten their potential shelf life, because unlike a block of wood or plastic, computers degrade over time.

This isn’t a problem for dedicated computers such as smartphones or computers, but consumer expectations around speakers, thermostats, and light bulbs are different. In the early days of the smart home — also when Sonos was introducing some of these products — it wasn’t something people thought about. But today, and even as far back as five years ago, what happens to a connected device after the company that makes it goes out of business or a customer needs to stop support was something people in the industry were talking about.

One way to address this problem is to mandate an expiration date for connected devices, and perhaps even a sell-by date. Doing so would make clear to the consumer that a product might not work after a certain date so they can then ascertain if the price is worth the benefit in the meantime. A sell-by date should also be printed on the device that gives notice to a customer that a device purchased after this date might not be supported for very long. The sell-by date helps in the resale market because a buyer could see that the device might only be “good” for another two years. Even today, I can go on Amazon and buy a connected device that’s unsupported and no longer receiving updates simply because Amazon isn’t policing resellers.

Also, in the case of the Sonos example, I personally resold an old Play 5 speaker on eBay last month, and now feel bad that I unknowingly sold someone something that will degrade in capability over time. If I had all the information up front, I either would have refrained from selling altogether, or I would have clearly disclosed what was likely to happen and let the buyer decide. Sonos’ response to this reseller conundrum is to “recycle” the devices by bricking them entirely so no one would ever be able to use them again.

I think the industry should do better by being more transparent. This isn’t an easy ask. It’s hard to know how long a company should plan to support something, and it will be tough to communicate an expiration date to customers without them rebelling. For example, my mom tries hard to avoid connected products (even in cars) because she despises the lack of longevity and user control. An expiration date wouldn’t change her mind; in fact, it would only steer her toward other products.

That is why such a date should be mandated by the government. Like privacy and security — other difficult topics for consumers to understand in depth — the lifecycle of connected products needs to be on everything so unscrupulous businesses can’t use a lack of an expiration date as a selling point, despite having no intention of supporting the product.

But an expiration date assumes that we want to continue down the path we’ve been heading — that is, with computing in every device and the focus on selling each product as a standalone item supported by one entity. An obvious option is to make dumber devices that communicate with a smart, connected device and attach the dumber product (a TV, a speaker) to a smart streaming device that might get replaced every three to five years.

Such an approach makes particularly good sense in the media streaming world, where services and codec can change frequently. It may also make sense for smart home hubs controlling lights and temperature. An added benefit would be that if a company goes under, they could release their code to the open-source community so that people could keep the devices going that way.

I’ve spoken to several entrepreneurs who feel there is a moneymaking opportunity in supporting such abandoned devices, but I’m not so sure. I welcome anyone to try it, though. Given the install base of the Sonos gear, the relative wealth of the buyers of such gear, and their degree of tech-savvy, it might work.

But I’d like to take a solution even further. One of the challenges associated with Sonos’ decision to leave its legacy products behind is that as the integrations between Sonos and other streaming services such as Spotify, Pandora, or Tidal change, customers will see their speakers stop working when they want to play music from those services.

But what if Sonos talked to those companies and acknowledged it was part of an ecosystem of media services that every company has a stake in? At that point, it might convince Spotify to support a legacy integration to keep customers of both services happy. Perhaps it would do this until only a fifth or a third of their mutual customers are left. One of the benefits of connected products is that a company has a really clear understanding of the number of users who would be affected and how much it might mean to them.

Sonos, for example, might have looked at its data and decided that only 20% of its customers have legacy products and of those, most stream from local storage, which means this announcement affects a small (but really vocal) minority. Given the type of data that connected products can provide, and the fact that most connected products are part of a web of other products and services, it seems that starting to approach obsolescence as an ecosystem makes sense.

If we’re using Sonos as a case study for how to think about what it means when our products become services, mandated expiration dates and hubs provide a good short-term solution, while a reimagined version of customer support shared among an ecosystem offers a long-term shift appropriate to the future we’re moving toward.