My First Week With a Continuous Glucose Monitor — Here's What Surprised Me

I'm not diabetic. I don't have a blood sugar problem I know of. And yet last Monday I stuck a coin-sized sensor into the back of my arm and let it read my glucose every few minutes for a week, purely out of curiosity about what my body actually does between meals. Seven days in, I'm still not entirely sure what to do with the data — but I'm already looking at food differently.

Continuous glucose monitors have quietly moved from "medical device for people managing diabetes" to "wellness gadget you can buy without a prescription," and I wanted to see what that shift actually feels like from the inside rather than reading another press release about it. So I ordered one of the over-the-counter sensors now aimed at people like me — nobody with a diagnosis, just someone curious about the numbers behind how I feel.

Applying it was stranger than I expected

The applicator looks like a chunky pen and works in one motion — press it against the back of your upper arm, hear a click, and a filament thinner than a hair sits just under the skin while a small sensor disc stays stuck on top. It didn't hurt going in, more a firm pinch than anything sharp, but I won't pretend the first thirty seconds of knowing there's a tiny wire under my skin weren't a bit unsettling. By the second day I'd completely forgotten it was there, except when a sleeve caught the edge of the disc.

The app took about an hour to start showing a proper graph rather than a flat "warming up" message, and from that point on I had a live glucose trace scrolling across my phone, twenty-four hours a day, whether I asked for it or not.

The first surprise: it wasn't the food I expected

I went in assuming the villain would be obvious — a pastry, a sugary coffee, the usual suspects. What actually spiked me hardest on day two was a bowl of plain porridge oats I've eaten most mornings for years without a second thought. The line on the graph climbed sharply, held, and came down slower than it did after a meal I'd have guiltily logged as "worse." Nothing about how that porridge tasted or felt going down gave any hint it was doing that. The graph was the only thing that told me.

I'm not about to swear off porridge based on one reading from a device with a healthy margin of error. But it planted a genuinely useful seed: my instincts about which foods are "fine" were built on taste and habit, not on anything my body was actually telling me.

The second surprise: sleep moved the line more than meals did

A bad night's sleep showed up on the glucose graph the next day almost as clearly as a meal did. After a late night and a rough five hours, my baseline sat noticeably higher the whole following day, and a lunch that had barely registered on a well-rested day produced a much sharper spike. I'd read that sleep and blood sugar were connected in the abstract, but watching it happen in my own data, twice in one week, landed very differently than reading a study about it.

Stress did something similar. A tense hour dealing with a work deadline nudged the line upward with no food involved at all — cortisol doing exactly what it's supposed to do, apparently, even when there's nothing to actually run from.

"I expected a food tracker. What I got, in week one at least, was closer to a stress and sleep tracker that happens to also notice food."

What the app first impressions actually looked like

The interface is clean and doesn't try to overwhelm you with clinical detail on day one — a live number, a trend arrow, and a rolling 24-hour graph, with more detailed logging available if you want to tag meals, workouts, or sleep against the line. Notifications for sharp rises can be switched off entirely, which I did by day three, because getting buzzed mid-meeting that my glucose was "trending high" after a normal lunch was more anxiety-inducing than useful.

Battery life on the sensor itself is a non-issue — it's sealed and disposable, rated to last the length of the wear period rather than needing any charging. Syncing to the phone over Bluetooth has been reliable, though I did lose a few readings the one time I left my phone in another room for an hour.

Where I'm already sceptical

• These devices are built and validated for people managing diabetes, where tight glucose control has clear, well-established stakes. For someone without that condition, the "normal" range the app flags things against is a general guideline, not a personalised medical target — worth remembering before treating every wobble on the graph as a verdict on your health.
• A week of data from one person proves very little. One porridge spike, one bad-sleep pattern — that's an anecdote, not a trend. I'd want a full month, across a proper range of meals and sleep patterns, before drawing any real conclusions.
• The marketing leans hard into "optimise everything," and it's easy to see how that tips into obsessing over a number that fluctuates naturally throughout any healthy day. I've already caught myself checking the app more than is useful.

What I'm doing differently already

I've swapped my usual porridge for a version with more protein stirred in, mostly out of curiosity whether it flattens the curve, and I've started treating a bad night's sleep as a reason to eat a bit more carefully the next day rather than a reason to reach for sugar to compensate. Neither is a dramatic life change. Both came directly from watching a graph I wouldn't have had any other way.

I'm going to keep wearing it for a full cycle before I write anything resembling a final verdict — this is a first-week reaction, not a review. If the porridge spike holds up, if the sleep pattern repeats, and if the novelty wears off without the anxiety creeping in, I'll come back and tell you honestly whether it earned a permanent spot on my arm.

If you want to see where this goes before it hits the site, join the Brite newsletter — no fluff, just what's actually worth your money.

Questions people actually ask

Does applying a continuous glucose monitor hurt?

Not really — it's a firm pinch rather than anything sharp. The applicator works in one click, leaving a filament thinner than a hair under the skin, and while the first thirty seconds of knowing it was there were a bit unsettling, I'd completely forgotten about it by day two apart from the odd sleeve catching the sensor disc.

Are CGMs accurate for people without diabetes?

Treat the numbers with caution. These devices are built and validated for people managing diabetes, and the normal range the app flags things against is a general guideline, not a personalised medical target. The readings carry a healthy margin of error too, which is why I'm not drawing conclusions from single spikes in my first week.

Is a CGM worth it if you're not diabetic?

One week in, I'm not ready to call it essential — the jury's still out on whether the insight justifies the cost and the low-grade anxiety of watching a number all day. That said, it has already changed how I think about foods I'd assumed were fine, which is more than I expected from seven days.

What raises blood sugar besides food?

In my week of data, sleep and stress moved the line almost as much as meals did. A rough five-hour night left my baseline noticeably higher the whole next day, and a tense hour on a work deadline nudged glucose upward with no food involved at all. That was the biggest surprise of the experiment.