Blog / smart home
Will Your Smart Home Still Work in Five Years?
Cloud shutdowns turn working smart home hardware into plastic. Here is the question to ask before you buy, and what to do with the gear you already own.
Belkin switched off the Wemo cloud on January 31, 2026. The plugs and switches people had installed did not physically change that day. The app that told them what to do stopped working, and for most owners that amounted to the same thing.
It is worth sitting with that for a minute, because it will happen again with a different logo on the box.
You only bought half the product
Most consumer smart home gear is two things sold as one. There is the piece of hardware you screwed to the wall, and there is a service running in a data centre that the hardware phones home to. You own the first part. You are borrowing the second part for as long as the manufacturer feels like paying the bill for it.
Nobody mentions this in the aisle at the hardware store. The box talks about voice control and scheduling and your phone, all of which quietly depend on somebody else's servers staying up and somebody else's app staying in the app store.
That arrangement is fine while the product is selling. It gets interesting five or six years later, when the line is discontinued, the company is acquired, or the recurring cost of running the cloud stops making sense to whoever now owns the balance sheet.
The question I would ask before buying anything
If the manufacturer disappeared tonight, what still works tomorrow morning?
There are really only three shapes of answer. Everything keeps working on your own network and you just lose remote access. The basics keep working (the switch still switches, the lock still locks with a code) but the clever parts go. Or the device is a paperweight. The third answer is more common than people expect, and it is the one that never appears in a product review.
You can usually work it out before you buy by looking for one thing: does the device do its main job when the internet is down? Pull the plug on your modem and try it. A light switch that will not turn on a light during an outage was never really a light switch.
Standards help, with caveats
This is the actual argument for buying into Matter, the interoperability standard run by the Connectivity Standards Alliance. Matter devices are designed to run locally on your own network, so their core functions do not depend on an internet connection or on one company's app surviving. The standard is still moving (version 1.6 landed in June 2026), and support varies by product category.
Two caveats I would not skip. A Matter logo covers the basics, not the manufacturer's extra features, so the fancy automations in the vendor app can still vanish while the on/off keeps working. And cameras are largely a separate problem: video is where the subscriptions live, and a camera that only records to somebody's cloud is the most fragile thing in the house. If a camera cannot record locally to an SD card or a recorder you own, assume it has an expiry date.
What to do with what you already own
Do not rip anything out. Working equipment that does its job is not a problem to be solved, which is a point I have made in more general terms before.
Do make a list. Every connected thing in the house, what it depends on, and whether it survives a dead app. Ten minutes with a notepad. The value is not the list, it is that you find out which items are load bearing. A cloud dependent smart plug on a lamp is an annoyance if it dies. A cloud dependent lock on the only door, or a thermostat in a cottage you heat remotely in February, is a different category, and those are the ones to replace on your own schedule instead of on the vendor's.
When you do replace something, prefer the boring option: hard wired where it makes sense, local control, a standard rather than a proprietary app, and no subscription for functions the hardware could perform on its own. Lighting is the clearest example, since a Lutron style system runs its schedules on a hub in your house and the dimmer on the wall still works as a dimmer regardless (my company installs that gear, so weigh that as you like). There is also a more involved version of this, where you run the hub yourself, which I have written about separately in whether Home Assistant is worth it. The boring option is also usually the one that does not embarrass you at closing time, which I have written about in the context of resale.
The wider point
This is not really about smart homes. It is the same question that decides whether a business platform, a hosting stack or a line of business application is safe to build on: what happens to me if the company behind this changes its mind? You cannot make that risk zero. You can decline to put it on the front door.
If you are planning a build or a renovation and want the wiring done so the automation is not the only thing holding the house together, my company RelayPoint Technologies does that installation work across Southern Ontario. If you would rather have someone look at what you already have and tell you what is fragile, that is an assessment, quoted as a fixed fee. Either way, get in touch and I will tell you honestly which of the two you need.