Chapter containing info on the KS3 from my forthcoming book 'Sugarland by Andrew Smales'. Available April. Scroll down in this window to read.

Chapter 5

When Algorithms Take Over - Measurement vs Marketing

Some technology quietly improves your life. Some of it quietly tries to run it. The difference isn’t whether it has an app or a graph. It’s whether it’s measuring reality - or using algorithmic nudges to steer behaviour.

The best health tech does one simple thing well. It takes something your body is already doing and shows it to you honestly: heart rhythm, oxygen levels, movement, sleep patterns, blood glucose, blood ketones and real blood pressure. It stays in its lane and lets you make the decisions.

A proper blood pressure cuff works because it’s honest - a pump, a valve and physics. It doesn’t necessarily need to be bluetooth or high-end even. A friend of mine bought one on Amazon for £10 and it works just fine. A continuous glucose monitor changed metabolic health because it measures reality instead of guessing it. Continuous ketone monitoring can do the same thing for fat-burning: not magic, not motivation, just a steady stream of data that makes cause-and-effect harder to dodge.

The Apple Watch built a health empire by using real sensors and, crucially, knowing its limits: tracking rhythm, flagging possible AFib, showing oxygen saturation, movement and sleep trends without pretending to diagnose everything.

That’s real tech. That’s measurement. And it’s been genuinely useful for me, even though it’s the ugliest thing to have ever sat on my wrist! Watching glucose flatten after meals. Seeing ketones rise when fat-burning is actually happening. Spotting when electrolytes are off because heart rate jumps. Checking oxygen levels during exertion. Catching rhythm weirdness early instead of guessing how I feel. Over weeks and months, those trends tell a real story.

But fat-burning was the one bit that still felt like folklore.

After the stroke, I’d already lived through one big plot twist. I’d made the grim train trip to London, sat in front of a giant screen, and watched a stranger point at my arteries and tell me my average blood sugar was in the “you could die from this” range. I came home, took the pills, changed the food, strapped on glucose monitors, and reversed the diabetes numbers in just over two months. I call it ‘reversed’ because the numbers normalised - but clinically it’s safer to say remission: it stays quiet as long as you don’t go back to the old inputs. In the process I also lost over 17 kg.

On paper it sounds almost neat - cut the sugar, cut the beige carbs, job done - but living it was months of grind, cravings and doubt. That gap between “it’s simple” and “it’s actually hard” is the whole reason I’m writing this book. If I can haul myself back from that edge, you can, too.

Glucose I understood. I could see the spikes, the crashes, the flat calm. I knew what toast did (you can’t have it). I knew what a walk did. I knew that cola and “healthy” cereal weren’t food, they were a graph with sirens.

But Keto has this other promise that’s harder to pin down. You’re told, “you’re burning fat now.” Great. How much? When? What kills it? What helps it? Am I actually in ketosis, or just hungry and grumpy with nice macros?

Finger-prick ketone tests helped, but they were like looking at a postcard once a day and trying to reconstruct the whole holiday. One reading in the evening doesn’t tell you what your “Keto dessert” did at 3 p.m., or whether that bone broth at nine o’clock kept the fire going or put it out.

What I wanted was the thing the stroke had forced on me in London: clarity. A screen that didn’t care how I felt about the story. Just the truth.

So I tried a continuous ketone sensor - the same basic idea as a CGM, but for ketones. You wear it, it updates in real time, and instead of turning your day into one heroic number, it shows you the line. And the first week I wore one, the guessing died. Not in a dramatic “new me” way. In a boring, practical way.

Coffee, sweetener, a short walk - I could see the line twitch in minutes, not days. The app didn’t nag or praise; it just showed me what was actually happening. I stopped arguing with myself and started steering: a little less sweetener here, a later meal there, adjusting when the curve drooped instead of waiting a month and wondering why my jeans felt tighter.

That’s the real power of measurement: feedback that lands before your willpower runs out.

It doesn’t burn fat for you. It doesn’t fix your diet. It doesn’t make you virtuous. It just makes the fire visible. Once you can see it, you stop doing dumb things that put it out.

It also does something else, which is less flattering. It catches you lying to yourself. One afternoon I’d eaten almost nothing - a 20 g piece of cheese and a swig of milk, that was it. By one in the morning I felt light, clear, properly “keto-ish”, the way influencers describe in their reels. I checked the reading, more out of curiosity than concern.

0.1.

Not heroic fat-burning at 5.0. Not even a jog. Just a flat, low line.

Before this, I’d have told that story as, “I hardly ate, I must have been in deep ketosis.” With the sensor on, that story died in one glance. Feeling Keto is not the same as being Keto. The sensor is the only one in the room that doesn’t believe my excuses.

Once you have that kind of feedback, you start running little tests - not because you’ve become a scientist, but because you’ve become allergic to superstition.

The “innocent” Keto dessert was the first one. Almond flour, a bit of sweetener, high calories, looked the part. The line blinked, dipped and flattened for hours. Next day, I skipped it. The line held, the morning felt better, and the scale finally nudged. Same intention, different outcome - because I could see the effect as it happened.

Bone broth on a fasting day was another. I kept the salt high, took the cup, watched the line barely flinch. A 20-minute walk after? The curve lifted like a kite in a sea breeze. That became a little protocol

for me: broth + salt + short walk = fire stays lit. Addictions hate evidence.

Then the zero-sugar drinks question, because everyone has an opinion and half of them are basically theology. Some people get a cephalic insulin nudge or cravings that push glucose up and ketones down - the idea of sweet can trip the body, even without sugar. So I did what I did with glucose: tested it. Empty stomach, one can, then another, watching the trend long enough to be sure. Mine didn’t budge. No dip, no delayed sag, no appetite spike. The line stayed flat and I got on with my day.

My rule from that point was calm and personal, not evangelical: for me, zero-sugar drinks are a taste, not a trigger. Your own body might disagree. That’s the value of the graph.

This is why continuous monitors - glucose and ketones - are in a different category from most wellness gadgets. They don’t “inspire” you. They don’t flatter you. They don’t sell you a personality. They just show you a number and let you deal with it.

Then there’s the other side. The wristbands that promise blood pressure without a cuff. The breath gadgets that promise metabolism from a puff of air. They don’t measure - they infer. They guess, smooth, model and extrapolate until a neat line appears on a screen and you feel like you’ve learned something.

Apple even does its own version of this with the so-called blood pressure hyperactivity “trend” in Health. It isn’t measuring blood pressure at all - it’s massaging months of unrelated data into a vague indicator and presenting it like insight. No cuff. No pressure reading. Just algorithmic guesswork wrapped in a reassuring graph. Nothing beats a real blood pressure measurement. Some things you just can’t improve. The moment tech stops measuring and starts modelling, you’re no longer looking at health - you’re looking at a story.

But inference isn’t measurement. It’s marketing dressed as biology. You end up trusting noise because it looks like science. You chase perfect graphs instead of real improvement. Before you know it, the gadget is running your day instead of your body.

Good tech listens. Bad tech performs. The best devices disappear into the background and the worst ones demand attention. Not every graph deserves your trust.

A quick honesty break, because trust beats theatrics.

The ketone sensor I use is made by SiBionics (the “SiBio” name is what most people call it), and I’m an authorised UK seller. The same with the LinX glucose monitor. If someone buys through me on eBay or ketogear.com, I benefit financially. That’s the disclosure.

But I’m not putting it in the book because I sell it. I sell it because it’s one of the very few gadgets I’ve used that actually pushed me closer to reality instead of further into vibes. It even tells you how much fat you’re burning per hour, which is a powerful motivator. If it didn’t work for me, it wouldn’t be here - because I’m not writing an advert, I’m writing a map out of Sugarland.

The Sleep Score (Useful, Until It Runs the Show)

I actually like sleep tracking. Over weeks and months it’s helped me spot patterns - late meals, alcohol, heat, stress, when I actually got to sleep - all of which show up clearly in the data. Used properly, it’s a brilliant feedback tool. Where it gets slippery is when a single night gets turned into a verdict. One morning I slept straight through, woke up feeling reasonably okay, made coffee - and checked my phone. Sleep score: sixty-two. Nothing in my body had changed in those five seconds, but suddenly I felt worse than I had a moment earlier. Not because I’d slept badly, but because I’d been told I had.

That’s the danger of compressing eight hours of messy human biology into one neat number. Sleep isn’t a score - it’s a conversation between hormones, light, stress, food, temperature, pain, noise and whatever your brain is processing from life. The app sees movement, heart rate and maybe oxygen, then estimates the rest.

Over time those estimates are useful. Over one night they can be misleading. So I made a simple rule: body first, graph second. When I wake up I check in with how I actually feel before I look at the data. If the two match, great. If they don’t, I trust the body.

Sleep tracking is powerful when it shows trends. It becomes a problem when a single number starts running your day.

Selling Air on Subscription

Then there’s Lumen - the breath gadget that claims to “hack your metabolism.” Strip away the marketing and it’s basically a carbon dioxide sensor. That’s it. A small CO₂ meter wrapped in slick branding.

It assumes your body should constantly flip between burning carbs and fat, which might make sense on a high-carb diet. On Keto - where consistency is the whole point - it simply breaks. Even Lumen admits ketogenic diets distort the results.

When I pointed this out to them, they didn’t fix it. They got defensive. Suddenly the conversation wasn’t about accuracy - it was about how I should be grateful for access to their “valuable proprietary software” and how unreasonable I was for expecting the hardware to work without a subscription. They had no issue with a paid-for device becoming a brick the moment you stopped paying.

What you’re really paying for isn’t measurement. It’s colourful charts, daily scores and the feeling of insight. And when the subscription stops? The expensive gadget becomes useless plastic. No recycling scheme. No take-back. Just e-waste. A basic blood ketone meter is cheaper, more accurate, and actually measures what matters. That’s Almost Science in action: real hardware, fake exaggerated insight.

Never Hand Your Health to a Company That Can Ghost You

Withings got me properly hooked at first. I didn’t just buy the scales - I bought the whole idea. Charts, trends, progress lines, the promise that data would finally keep me honest. I preferred the classic watch design to Apple’s rather ugly square one. It felt like smart health tech done right, so I kept adding to it. The ScanWatch Nova at £549. A strap for £80. Body Comp scales at £215. The BPM Core blood pressure monitor at nearly £300. A sleep analyser mat for £130. Even the U-Scan urine sensor at almost £400.

By the time I added it up, I’d spent well over fifteen hundred pounds building what felt like a personal health lab.

Some of it was genuinely clever. The sleep mat tracked breathing and movement without wearing anything. The blood pressure monitor worked properly because it used a cuff. The watch tracked rhythm and activity well, and had a three week battery life.

The urine sensor sounded like the future - daily biomarker tracking without needles or strips. In reality it barely measured anything useful. Simple urine test strips were clearer, cheaper and far more informative. It was tech for the sake of tech.

Then Apple updated iOS. And everything broke. Every update knocked the Withings app over.

Devices stopped syncing. Data vanished. For weeks nothing worked properly. After a month of chasing support they admitted the software wouldn’t be fixed for another month.

That’s when it hit me: every piece of hardware I’d bought was completely dependent on software I didn’t control. I refused to wait indefinitely while thousands of pounds of equipment sat useless. A product that only works when an app behaves isn’t ownership - it’s rental.

To their credit they refunded the hardware. But it took nearly four months, and it was done grudgingly rather than cleanly. It was a shame - because the tech had real potential.

But the lesson stuck. Data is useful. Trends are useful. But your health should never be hostage to updates, subscriptions or corporate timelines. Tech should support your body - not sit between you and it.

When Algorithms Manufacture Cravings

The same logic doesn’t just live in health tech.

One night I deliberately left an Uber Eats notification on my Apple Watch. It buzzed and suggested food I hadn’t thought about seconds earlier. I wasn’t hungry until the algorithm told me I should be. That’s modern Sugarland: manufacture the craving, then sell the cure.

Ignoring that buzz felt like winning a level in a video game - ridiculous, but revealing. When resisting software feels like discipline, you’re not fighting hunger anymore. You’re fighting design.

Big Food plus Big Tech doesn’t wait for appetite. It creates it

The Rule That Keeps This Honest

I’m not anti-tech. I’m anti-nonsense.

Good technology measures reality and stays in its lane. Bad technology guesses, prettifies and sells you confidence instead of truth. Some tools quietly help you listen to your body. Others replace your body with an algorithm.

Use the tools. Keep your hands on the wheel. You’re not here to build a prettier dashboard. You’re here to get healthier in real life.

Author: © Andrew Smales 2025