What a CGM Costs, and What You Get for the Money

If you are paying cash and have no diabetes diagnosis, expect roughly $80 to $110 per month for an over-the-counter continuous glucose monitor (CGM), which works out to about $2.50 to $3.70 per day of data. Prescription sensors bought through insurance usually land between $0 and $60 per month in copays when the plan covers them, and $150 to $350 per month at cash retail when it does not. The spread is large because three different pricing systems (over-the-counter retail, pharmacy benefit, and durable medical equipment) sell nearly the same silicon and enzyme chemistry.

A CGM is a small filament inserted into interstitial fluid under the skin that measures glucose electrochemically every one to five minutes for 10 to 15 days, then is discarded. The sensor is a consumable, so the meaningful unit of cost is dollars per day of data, and the meaningful question is how many days you need.

The over-the-counter tier set the price floor

In 2024 the FDA cleared the first CGMs for sale without a prescription to adults not using insulin, and that tier now defines the low end of the market. Dexcom’s Stelo and Abbott’s Lingo and Libre Rio all sell direct-to-consumer, typically around $89 to $99 per month on subscription for two sensors, with single sensors near $50 and multi-month bundles cutting perhaps 10 to 15 percent off the monthly rate. Prices move, so verify at the vendor before ordering. The shape has been stable at two sensors for roughly $90, covering 28 to 30 days.

These are the same sensing platforms as the prescription devices with the software constrained. Stelo is built on Dexcom G7 hardware but reports on a 15-minute smoothing cadence and does not alarm for hypoglycemia. Lingo is Libre 3 hardware with a consumer app that computes proprietary “glucose spike” counts. If you intend to do your own analysis, that constraint matters more than the price difference, because it determines the temporal resolution of what you can export.

Prescription and insurance routes

Prescription CGMs include the Dexcom G7, FreeStyle Libre 3 Plus, Medtronic Guardian and Eversense. They are covered by most commercial plans and by Medicare for people who use insulin or have documented problematic hypoglycemia. Coverage is adjudicated either as a pharmacy benefit or as durable medical equipment, and the same sensor can differ by more than $100 a month depending on which bucket your plan uses, so it is worth asking the pharmacist to run it both ways. Warehouse-club pharmacies and discount card pricing (GoodRx and similar) generally bring cash Libre 3 Plus to somewhere near $130 to $180 for a month of sensors and G7 to roughly $180 to $250, which is the relevant number if you have a prescription but no coverage.

The cost-effectiveness literature explains why payers cover this at all. In the DIAMOND randomized trial of adults with type 1 diabetes, CGM compared with self-monitoring of blood glucose produced incremental quality-adjusted life-year gains at a cost that fell within conventional willingness-to-pay thresholds over a lifetime horizon.¹ Modelling in type 2 diabetes reached similar conclusions, driven mostly by avoided complications rather than by anything that happens in the first year.² In pregnancy complicated by type 1 diabetes, modelling of the CONCEPTT trial data suggested real-time CGM reduces neonatal intensive care costs enough to offset much of the device cost within a single pregnancy.³ None of that evidence applies to a healthy engineer buying a sensor to look at their own postprandial curves, and it is fair to say the health benefit in that population is unproven.

Cost per day of useful data

We would not wear a CGM continuously for self-measurement. The information content of the tenth consecutive month is low, because your glucose response to your usual breakfast is a stable property you can measure once. The approach we recommend is two to four 14-day blocks per year, each one paired with a disciplined food and activity log, at a total cost of roughly $200 to $400 annually.

Structure each block as an experiment rather than as ambient monitoring. Spend the first two days establishing your habitual baseline without changing anything, then run repeated standardized challenges: the same meal at the same time on three separate days, which gives you a within-person estimate of day-to-day variance before you compare foods. Without that replication, a 30 mg/dL difference between two meals tells you almost nothing. The same meal eaten twice can differ by that much depending on sleep, prior exercise, and sensor day.

Getting the raw data out

Every vendor app will show you a chart. What you want is a time series you can join against everything else you measure. Dexcom’s Clarity portal exports CSV with columns for timestamp, event type, and glucose value in mg/dL, at five-minute epochs, and the Dexcom Developer API (OAuth 2.0, with a sandbox environment and /v3/users/self/egvs for estimated glucose values) returns the same data as JSON with device and transmitter metadata. Abbott’s LibreView exports CSV from the account portal, with Libre 3 recording at one-minute resolution internally though the export cadence depends on app version. Stelo and Lingo data are exportable through their apps, but the granularity is coarser than the underlying hardware supports, so check what you get before committing to a platform.

A few practical notes on the files. Timestamps are local time without offset in most Dexcom exports, so daylight saving transitions will silently create a duplicated or missing hour that breaks naive resampling. Gaps appear wherever the phone was out of Bluetooth range and the sensor’s backfill buffer (three hours on G7) expired. Rows mixing calibration events, scan records, and estimated values share a column. Filter on event type first. For analysis, the iglu R package implements the consensus metrics with documented definitions. That is preferable to hand-rolling MAGE from a blog post, since at least four incompatible MAGE definitions circulate. The package covers the following metrics.

• time in range
• coefficient of variation
• MAGE
• CONGA
• J-index
• GRI

Where the money is wasted

Sensor failures are the largest hidden cost, and vendors replace them if you file promptly, so keep the lot number. The common failure modes are worth knowing in advance. Day one readings are noisy while the wound response settles, so many users discard the first 12 to 24 hours. Compression lows, caused by lying on the sensor, produce a smooth 40 to 60 mg/dL overnight dip that resolves when you roll over and are the single most common cause of alarming false readings. Adhesive failure in the second week is frequent enough that an overpatch is worth the two dollars.

Accuracy sets a floor on what any amount of money buys. Manufacturer-reported mean absolute relative difference is about 8 percent for both Dexcom G7 and FreeStyle Libre 3 against laboratory reference, which means a displayed 100 mg/dL is consistent with a true value in the low 90s or high 100s. Interstitial glucose also lags plasma by roughly 5 to 15 minutes during rapid change, so peak heights after a fast-absorbing meal are systematically blunted and shifted later. Ascorbic acid supplementation can bias Libre sensors upward. Independent accuracy studies in field conditions, including supervised pediatric camp settings, report performance close to but not identical to the controlled in-clinic numbers.⁴

If your readings suggest fasting glucose consistently above the normal range, or you see unexplained lows, that is a conversation with a physician and a venous lab draw. It is not a self-directed experiment. CGM is a measurement tool, and the clinical interpretation of abnormal glucose belongs with a clinician.⁵

Questions people also ask

How much does a CGM usually cost? Over-the-counter sensors run about $89 to $99 per month cash. Prescription sensors are $0 to $60 per month with typical commercial coverage and $150 to $350 per month without it.

How much does the FreeStyle Libre 3 cost at Costco? Costco pharmacy pricing with a prescription and the member prescription program has generally been among the cheaper cash options, often in the $120 to $170 range for a month of Libre 3 Plus sensors. This varies by state and changes frequently, so call the pharmacy with your prescription in hand for a current quote.

Is a CGM covered by insurance? Usually yes if you use insulin or have documented hypoglycemia, and usually no if you are asymptomatic and want one for general metabolic curiosity. Coverage decisions rest on the evidence base in diagnosed diabetes, where cost-effectiveness has been demonstrated in randomized trial data.¹

What is the cheapest price for Dexcom G7? For most people it is Stelo, which uses the same sensing hardware at roughly half the cash price of G7, with the tradeoff of coarser reporting and no hypoglycemia alarms. If you need alarms or insulin-dosing labeling, you need G7 and a prescription.

Is it worth getting a CGM? For a technically fluent person who will log meals carefully and analyze the exported CSV, two weeks of data is one of the higher-information-per-dollar measurements available in consumer health. Wearing one indefinitely without a specific question is where the money stops buying information.

Footnotes

1. Wen Wan, M. Reza Skandari, Alexa Minc, et al. Cost-effectiveness of Continuous Glucose Monitoring for Adults With Type 1 Diabetes Compared With Self-Monitoring of Blood Glucose: The DIAMOND Randomized Trial. Diabetes Care, 2018. https://doi.org/10.2337/dc17-1821 ↩ ↩²

2. Stephanie J. Fonda, Claudia Graham, Julie Munakata, et al. The Cost-Effectiveness of Real-Time Continuous Glucose Monitoring (RT-CGM) in Type 2 Diabetes. Journal of Diabetes Science and Technology, 2016. https://doi.org/10.1177/1932296816628547 ↩

3. H. R. Murphy, D. S. Feig, J. J. Sanchez, et al. Modelling potential cost savings from use of real‐time continuous glucose monitoring in pregnant women with Type 1 diabetes. Diabetic Medicine, 2019. https://doi.org/10.1111/dme.14046 ↩

4. Tatsuya Kondo, Takafumi Senokuchi, Mayumi Yano, et al. Benefits and Accuracy of Real-Time CGM in Children and Adolescents With Type 1 Diabetes Attending Summer Camp: Second Year. The Journal of Clinical Endocrinology & Metabolism, 2025. https://doi.org/10.1210/clinem/dgaf160 ↩

5. Mark R. Burge, Stephen Mitchell, Alison Sawyer, et al. Continuous Glucose Monitoring: The Future of Diabetes Management. Diabetes Spectrum, 2008. https://doi.org/10.2337/diaspect.21.2.112 ↩