Continuous glucose monitors were built for people whose lives depend on the number. Around 2020 a set of consumer platforms began prescribing the same sensors off-label to people with no diagnosis, wrapping them in apps that turn a glucose curve into food advice. The hardware works well. The interpretive layer on top of it is where the category gets complicated, because a signal designed to prevent hypoglycemia in a type 1 diabetic is now being used to decide whether someone should eat oats.
The buyer's real problem is that the sensor is a commodity and the subscription is not. Every consumer platform runs the same two manufacturers' hardware, so what you are choosing between is the app, the coaching, and the price — and, increasingly, whether you need a platform at all now that over-the-counter sensors exist. This hub covers what the device actually measures, how to read your own data without overreacting, and how long the value lasts.
The Verdict
The options, and what actually differs between them
Note the "sensor" column. The underlying hardware is manufactured by Dexcom or Abbott in every case, and its accuracy does not change based on which subscription delivered it. Listed alphabetically, with the direct-purchase route last because it is not a platform.
| Option | Typical price | Sensor used | What you are paying for | Best fit |
|---|---|---|---|---|
| Levels | ~$199/mo (sensors included) | Dexcom or Abbott, depending on supply | Consumer app, food logging, metabolic scoring | Users who want a polished app and self-directed learning |
| Nutrisense | ~$250–$400/mo by tier | Dexcom or Abbott, depending on tier | Registered dietitian consults included | Users who want a human interpreting the data |
| Signos | ~$140–$160/mo | Dexcom | Weight-loss framing and activity prompts | Users focused on weight management |
| Direct pharmacy purchase (over-the-counter sensor) | ~$50–$100 per 14–15 day sensor | Abbott Libre or Dexcom OTC line | No app layer beyond the manufacturer app | Users who want raw data without a subscription |
That produces the central cost question in this category. A subscription platform at $199 per month costs roughly $2,400 a year. Buying over-the-counter sensors for two separate two-week blocks costs somewhere around $200 for the year. The difference buys app design, food logging, scoring, and in some cases a dietitian. Whether that is worth ten times the price depends entirely on whether you would actually interpret raw data on your own.
What a CGM measures for someone without diabetes
For a diabetic, a CGM prevents dangerous highs and lows. For a non-diabetic, none of those events are in play, so the device is doing something different: revealing which specific meals, sequences, and behaviors move your glucose, in your body, right now.
That is genuinely useful information, because glucose response to identical food varies substantially between people. Two adults eating the same bowl of rice can produce meaningfully different curves. Population nutrition advice cannot capture that; a two-week sensor can. It also makes several non-food effects visible in a way nothing else does — a poor night's sleep raising the next morning's fasting glucose, alcohol suppressing overnight glucose then rebounding, a ten-minute walk after dinner flattening a peak by 20–30 mg/dL.
How to read your own curve without panicking
Most first-time users overreact to peaks. These are the reference points worth holding, with the caveat that individual readings carry meaningful error.
| What you are looking at | Reasonable range | What it suggests | Common misreading |
|---|---|---|---|
| Fasting glucose (on waking) | 70–90 mg/dL | Above 100 repeatedly suggests early insulin resistance | A dawn rise of 10–20 mg/dL is normal cortisol physiology, not a problem |
| Peak after a meal | Under 140 mg/dL | Repeated peaks above 160–180 in a non-diabetic are worth investigating | One high reading after a birthday cake means nothing |
| Return to baseline | Within 2–3 hours | Slow return matters more than peak height | Fiber, fat, and protein legitimately extend the curve without harm |
| Average glucose over 14 days | 85–105 mg/dL | The most stable number a consumer CGM produces | Compare it against your HbA1c — large disagreement suggests sensor drift |
| Standard deviation / variability | Under ~15 mg/dL | Flatness is not the goal, but wild swings track poorly with metabolic health | Exercise causes sharp rises that are entirely healthy |
Two patterns get misdiagnosed constantly. The first is exercise: hard training raises glucose sharply, because the liver releases stored glucose to fuel the work. That spike is healthy and should not be avoided. The second is the compression low — sleeping on the sensor restricts local blood flow and produces a false reading that can appear as a dramatic overnight drop. If your curve shows a plunge at 3 a.m. only on nights you slept on that arm, the sensor was compressed, not your blood sugar.
The insight plateau — and why subscriptions outlive their value
CGM value is front-loaded and it drops off sharply. The learning curve runs roughly like this. Week one is calibration to the novelty, and most of the data is discarded once you understand what you are looking at. Weeks two through four are where the real findings land: the three or four regular meals that spike you hardest, the effect of meal order and post-meal walking, and what alcohol and short sleep do to your fasting number. Weeks five through eight test fixes and confirm they work.
After that, the curve mostly repeats. You are re-observing conclusions you already drew, at $150–$400 per month. Some users genuinely need continued monitoring — people actively reversing prediabetes, people on GLP-1 medications watching their response, or athletes managing fueling. For everyone else, a defensible pattern is one to two months of wear, act on what you learned, then a repeat block six or twelve months later to check whether the changes held.
Does a CGM help with weight loss?
A CGM does not cause weight loss, and the randomised evidence for it as a weight-loss tool in people without diabetes is thin. Trials that have compared CGM-guided eating against conventional dietary advice in non-diabetic adults have not shown a consistent advantage for the sensor. That is the honest state of the evidence, and it sits awkwardly beside how the category is marketed.
One mechanism is real, though, and it is not the one the advertising emphasises. Seeing a curve rise after a specific food produces immediate, personal feedback of a kind that a calorie target cannot — and adherence is the variable with the strongest evidence behind it in every dietary intervention studied. If watching your own response is what finally makes you stop eating a particular breakfast, the sensor helped. It helped through behaviour change, not through metabolism.
The failure mode worth naming is the opposite one. Treating every post-meal rise as damage pushes people toward eliminating carbohydrate sources that were never the problem, including fruit, legumes, and whole grains. A rise to 120–140 mg/dL that returns to baseline within two to three hours is normal physiology in a healthy adult. Chasing a flat line is not a weight-loss strategy and, for some users, becomes an anxious relationship with food that outlasts the subscription.
The prescription pathway, step by step
There are now two routes, and the difference is worth understanding before you subscribe.
- Through a consumer platform. You complete an intake questionnaire, the platform's affiliated telehealth network issues a prescription, and sensors ship on a subscription cadence. Your own physician is not involved, and the prescription is tied to continued enrollment.
- Over the counter. Since 2024, continuous sensors intended for adults not on insulin have been available without a prescription, bought directly from a pharmacy or the manufacturer and paired with the manufacturer's own app. No subscription, no telehealth intake, and you can buy exactly as many sensors as you plan to wear.
Insurance almost never covers either route for a non-diabetic user. Health savings and flexible spending accounts sometimes do, and that is worth checking before paying out of pocket — it is effectively a 20–35% discount depending on your tax situation.
Reviews
- Levels review — polished consumer app, subscription-heavy
- Nutrisense review — includes dietitian consults
Explainers
- CGM accuracy — MARD, interstitial lag, compression lows, and false readings
- Over-the-counter CGMs — Libre Rio, Lingo and Stelo without a prescription
Related
- Biomarker guides — fasting insulin, HbA1c, and what glucose alone misses
- Reversal vs management — where continuous glucose data genuinely earns its cost
- Nutrition guides
- Wearables compared
Frequently Asked Questions
What does a CGM actually measure?
Not blood glucose. A CGM measures glucose in interstitial fluid — the fluid between cells just under the skin — using a filament inserted a few millimeters deep. Interstitial glucose lags blood glucose by roughly 5–15 minutes and is estimated from an electrochemical signal, not read directly. That lag is why a CGM reading during a fast rise or fall will disagree with a fingerstick, and why neither is wrong.
Are glucose spikes bad?
Spikes are normal physiology. A healthy non-diabetic adult eating a meal with carbohydrate will see glucose rise, often to 120–140 mg/dL, and return to baseline within two to three hours. That is the system working. What tracks with metabolic dysfunction is the pattern: repeated peaks above 160–180, a return to baseline that takes longer than three hours, or a fasting level that will not settle below 100. Judging a single reading in isolation is the most common way people misread their own data.
How accurate is a consumer CGM?
Modern sensors run a mean absolute relative difference of roughly 8–10% against lab reference values, which is good but not precise. In practice, two sensors worn simultaneously on the same person can differ by 10–20 mg/dL, and the first 12–24 hours after insertion are the least reliable. Treat the shape of the curve as the signal and any individual number as approximate.
Do I need a prescription for a CGM?
It depends on the product. Consumer subscription platforms route you through their own telehealth network, which issues the prescription as part of onboarding — you fill out a questionnaire rather than see your own doctor. Since 2024 there are also over-the-counter continuous sensors available without any prescription. If you only want the data, the over-the-counter route removes both the prescription step and the subscription.
How long should a non-diabetic wear a CGM?
Most of the learning happens in the first 30–60 days. In that window you find which of your regular meals spike you, how much a walk after eating flattens the curve, what poor sleep and alcohol do to your fasting number, and how your body handles the same food at different times of day. After that you are mostly re-confirming known results. Two to four weeks of sensor time, repeated once or twice a year, captures most of the value at a fraction of the annual subscription cost.
Should I use a CGM instead of blood work?
No — they answer different questions, and blood work answers the more important one first. A CGM shows glucose behavior in real time but says nothing about how hard your pancreas is working to produce it. Fasting insulin and HOMA-IR reveal insulin resistance years before glucose patterns look abnormal. If you can only afford one, run the blood panel. A CGM is most useful layered on top of a panel that already flagged something.