There’s a category of purchase you make when you’re tired enough, and I’ve made most of them. A watch that scores my nights out of 100. Supplements by the drawerful. A list of “sleep hygiene” upgrades long enough to need its own spreadsheet. I have fibromyalgia, a chronic illness that wrecks sleep as a matter of routine, so I’ve had more reason than most to spend my way toward a better night. What I’ve learned, slowly and at considerable cost, is that the sleep market has the value equation almost exactly backwards.

The strongest evidence is mostly free and pretty boring, because the things that cost the most and generate the most data are the ones most likely to make an anxious sleeper worse. If you’re shopping for sleep right now, that’s the single most useful thing I can tell you before you spend anything.

I’ll start with the wearables, because they’re where most people begin and where the money starts to compound.

A sleep tracker is good at one thing and shaky on others, and the marketing tends to lean on those dodgier areas. When researchers put six wrist-worn devices to the test against polysomnography, the in-lab gold standard, all of them detected more than 90% of sleep epochs but showed much lower specificity, resulting in only fair-to-moderate agreement on which stage you were actually in.

In plain terms: they can tell roughly when you’re asleep, and they struggle with the deep-versus-light-versus-REM breakdown that the app presents as a tidy coloured chart every morning. A separate multicentre validation of 11 trackers found accuracy scores ranging from decent to genuinely poor across devices, suggesting “sleep tracker accuracy” isn’t a single thing you can trust or distrust as a category. It varies based on hundreds of pounds and by brand. The Oura ring tends to come out best on stage detection; plenty of others land at moderate (at best).

None of that would matter too much if the chart were just a curiosity for people. The problem is what the chart does to people who care about it, and there’s now a name for it. Orthosomnia describes an unhealthy preoccupation with achieving “perfect” sleep as defined by your tracker, and it can push you to optimise for the metric at the expense of the thing it’s supposed to measure in the first place. Lying still in bed awake, for instance, improves the data while doing very little for your well-being.

Dr Sabra Abbott, the Northwestern neurologist who helped coin the term, models it explicitly on orthorexia, the unhealthy fixation with eating “correctly.” Roughly a third of people now use a sleep tracker, and a 2024 study of 523 adults found that tracker users scored higher on insomnia symptoms than non-users. That isn’t proof that the tracker caused it, and the researchers are careful on that point. But the direction of travel is hard to ignore, and the mechanism is intuitive enough: strap a scoreboard to your wrist and the bed suddenly becomes another place where you’re failing.

For someone already prone to anxiety (and I’d put my hand up here), that’s not a neutral gadget. It’s a nightly performance review you’ve paid to receive.

Which brings me to the premium end, where the price tags get more serious, and the questions get a bit more important.

The smart mattress category, meaning Eight Sleep, Sleep Number, BedJet and the rest, is built around active temperature regulation, and here the evidence is genuinely better than the tracker hype would lead you to expect. An Eight Sleep Pod cover costs around £2,000 or more and circulates water through it to heat or cool each side of the bed. Tom’s Guide, whose testers do this for a living, rate the climate control as the thing they’d actually keep. Cooling the sleep surface has real evidence supporting faster drop-off, particularly if you run hot.

The honest caveat is the one most brand pages skate past. As one in-depth review puts it, the benefit is strong for hot sleepers and minimal if you already sleep at a comfortable temperature. Without the ongoing paid membership, the clever cover reverts to a fairly dumb heating pad, losing the automation and tracking you bought it for. So the real question isn’t “is it good.” It’s about whether you’re the right person for it, and whether you’re comfortable paying a subscription on top of two grand to keep the smart bits switched on.

If you sleep hot and have the budget, I’d say it’s worth it. If you’re buying it because your sleep score is bad and this is the next thing to try, I’d think twice.

Wake-up lights are the one bit of bedroom tech I’ll defend almost without hesitation (you can see how I got on with a bunch for The Independent), partly because they’re pretty cheap and partly because they’re honest about what they do. A dawn simulator brightens gradually before your alarm, and the clinical evidence suggests it cuts the severity and duration of sleep inertia, that groggy, stuck-in-mud feeling, by roughly half on average. The effect on daytime alertness is real but modest and a bit mixed across studies, and it’s clearer for winter-mood dips than for turning anyone into a morning person.

The same page has the line that should manage your expectations before you buy: a dawn simulator can wake you, but it won’t get you out of bed. At £30 to £130, though, the downside is small(ish), and the gentler wake is a genuine daily pleasure.

Here’s the part the market has no real incentive to sell you, because there’s almost nothing to sell.

The first-line treatment for chronic insomnia isn’t a device at all. It’s cognitive behavioural therapy for insomnia, recommended ahead of medication by the American Academy of Sleep Medicine, the American College of Physicians and the British Association for Psychopharmacology, and shown to outperform sleeping pills over the long term without dependence or rebound. App-based versions exist; some are free. None of it photographs well for a glitzy launch event.

And the most impactful single change, according to the people who treat this for a living, is the least glamorous thing imaginable: a consistent wake time, every day, regardless of how the night went. It anchors your circadian rhythm and builds the sleep pressure that makes the next night easier. Round that out with a cool, dark room of about 16–18°C, caffeine cut after midday, and not using alcohol to nod off, and you’ve covered the evidence-backed basics. Total cost: nothing, beyond the discipline of getting up at the same time on a Sunday when you’re just the right amount of cosy.

I find this genuinely annoying, for the record. I would love for the answer to be a purchase. A purchase is a clean, satisfying action you can take at 11pm when you’re wired and despairing, and a fixed alarm on a bad morning is neither. But the gadgets cluster at the wrong end of the effort-to-payoff curve: the more a product costs and the more data it produces, the more likely it is to feed the anxiety rather than the sleep.

If you’re tired and tempted to spend, the order I’d actually buy is the reverse of the marketing. Fix the wake time first, because it’s free and it works. Get the cheap light if dark mornings flatten you. Consider the expensive mattress cover only if you genuinely sleep hot and your budget allows. And the tracker — the thing most people reach for first — I’d leave until last, if at all, and the moment checking the score becomes a morning ritual that sets the tone for the day, that’s your cue to take it off.

The best sleep purchase I ever made was the decision to stop measuring. Which is a strange thing to write at the bottom of a shopping column, but possibly the most useful.

I’m a freelance journalist and content strategist. Available for commissions and content-strategy projects across smart home, homes, lifestyle, consumer tech, B2B and entertainment. carolinepreece89@gmail.com

I have three security cameras in a one-bedroom flat in Ipswich, and not one of them points at the front door. They’re review units mostly, gathered over the years I’ve spent testing smart-home kit, and they’ve drifted into a single purpose: finding a black cat who can fold himself into a dark cushion and sleep there, undetected, for most of an afternoon.

His name is Jet, he’s my first cat, and because he’s indoor-outdoor and treats the neighbours’ gardens as his bathroom, I’ve never had a reason to buy a smart litter tray. What I do own, and what I genuinely couldn’t give up, is the Tractive GPS collar around his neck.

The collar itself is cheap; the subscription runs somewhere between £80-£90 a year, and I pay it without thinking, because the alternative is standing at a rain-speckled window at midnight, wondering whether he’s shut in the shed next door. Love makes us very cooperative consumers. The pet-tech industry has worked that out and has been moving the goalposts ever since.

When I bought the Tractive, it answered one question: where is he? Over the past year of software updates, though, it’s stopped being content with coordinates. The app now hands me daily Wellness Scores, sleep-disruption alerts and an estimate of how many calories he’s burned chasing things in the garden I’ll never see (birds, I assume, and the enormous rat that sometimes materialises on the patio).

None of it is information I asked for. All of it nudges a situational safety device a little closer to being yet another health dashboard I’m meant to keep checking twice a day.

If your cat is indoors, the same shift is happening via the litter box. The latest smart models, like the Whisker Litter-Robot 4 at about £699 or the PETKIT Purobot Max Pro 2 at £470, have left behind the old infrared sensors. Now they use load cells to weigh the cat to the gram, cameras set low to the ground, and computer vision to measure exactly what gets left behind.

Whisker’s WasteID will tell urine from faeces (nice); both systems claim to recognise individual cats by face in a multi-cat house, where weight alone tends to fall apart once two animals are a similar size.

It would be easy to be sniffy about all this, and I don’t want to be, because the medical argument is genuinely strong. Cats are very good at hiding illness; by the time one is visibly in pain or off its food, conditions like feline lower urinary tract disease, a blockage or early kidney decline are often well advanced.

A box that tracks weight, visit frequency and time spent inside can spot a problem days before it turns into a late-night trip to the emergency vet. If you’re looking after an older cat with kidney issues, I’d take that early warning, and the £699 would feel like a small part of the deal.

The difficulty comes when that medical baseline meets the usual smart-home business model.

Whisker now charges roughly £8 per month, or £80 per year, for Whisker+ on its newer models. Skip it, and you’re capped at about five minutes of live viewing a day and two days of history. The long-term weight charts, the behaviour logs and the 2 years of trend data that the box has been collecting all require a subscription. You’ve bought a £699 appliance, and the thing it was sold to do, spot a slow drift in your cat’s health, is the thing that is held back unless you keep paying.

The question is no longer whether an automated scoop is worth the money (which you can answer), but whether cancelling to save £8 a month makes you a bad owner. It’s a tax on guilt, and it works. Cancel a streaming service, and you lose a few shows. Cancel this, and you start to wonder if you ever cared about your cat’s health in the first place.

There’s also a less glamorous problem: where do these things live? A video doorbell on the porch or a smart speaker in the kitchen is a compromise most of us have made with little trouble. A litter box, by necessity, often sits in the bathroom, the utility cupboard or the bedroom, the few rooms left where people still expect some privacy.

To keep an eye on a cat’s urinary health, you’ve put an internet-connected camera, usually with a microphone, at ankle height next to your own toilet, feeding a server whose security parameters you’ll never truly know.

On Reddit, a steady trickle of owners are talking about taking premium boxes off the network entirely, or refusing firmware updates, to stop exactly that. They paid the better part of a grand for the hardware and decided the surveillance wasn’t worth it. Hard to argue, really.

The promised peace of mind can quickly turn. Put the box on an uneven tile and the weight reading shifts; in a two-cat house, a misread visit triggers a generic alert. For an anxious owner, there’s perhaps nothing worse than a stream of medical data you can’t really interpret. I’ve watched a barely-moving spot on the map at 2am, not sure if it’s time to worry or if the cat’s just lost his collar (always use a breakaway!). I’m up at that time more often than I’d like, so the app and I have that in common. Living with an animal can mean trading real uncertainty for constant paranoia.

Most evenings Jet comes back from his rounds, jumps up on my chest, and spends twenty minutes washing his paws, completely indifferent to the cellular radio on his throat or the cameras that spent the afternoon hunting for him. He runs on instinct and has no opinion about his own longevity.

But the industry has noticed something: cats don’t care about data, but their owners care about cats long past the point where it makes financial sense. For some animals and some conditions, early warning is a genuine improvement, and I wouldn’t ask anyone to give that up. But as more of these devices find a place in the house, we’re being asked, one payment at a time, how much of the home we want to wire up to answer a question we once never even considered.

I’m a freelance journalist and content strategist. Available for commissions and content-strategy projects across smart home, homes, lifestyle, consumer tech, B2B and entertainment — the useful version, not the brochure one. carolinepreece89@gmail.com