App review
Diabetes Management Apps in 2026: What Each Layer Actually Does
There is no single diabetes management application, and the products claiming to be one are usually weakest at the layer that matters most. Management divides into four: the readings, which belong to your sensor or meter's own regulated application; the treatment record of insulin and medication, where mySugr is purpose-built; the food record that explains why the readings moved, where PlateLens leads on the only independently replicated accuracy figure in the category and logs glucose free; and the clinical review, which is a person rather than software. This page ranks the third layer, states plainly what belongs in the first two, and explains why an all-in-one product tends to be a compromise at the point of greatest consequence.
At a glance
| Best for | People with type 1 or type 2 diabetes assembling a set of tools, and anyone wondering why no single app seems to do the whole job. |
|---|---|
| Pricing | Manufacturer applications are included with the device. Third-party applications range from permanently free tiers to roughly $80 a year. |
| CGM integration | Dexcom G7, FreeStyle Libre 3, Apple Health, Health Connect |
| FDA status | Sensors, meters and their manufacturers' applications are regulated devices. None of the third-party applications discussed is FDA-cleared as a medical device, none contains a regulated bolus calculator, and none should be used for treatment decisions. |
| Carb-accuracy score (editorial) | 9.3 / 10 · composite of validated MAPE evidence (where available), database provenance, and clinical workflow fit |
Strengths
- Separates the four layers instead of pretending one product covers them
- States which functions are regulated and must not be delegated to a third-party app
- Reviewed by the medical reviewer; regulatory status given for every application named
Limitations
- Editorial rankings rather than a single in-house quantitative protocol
- Insulin dosing algorithms and pump integration are out of scope
- No application named is a medical device or a dosing aid
“Diabetes management app” describes four different jobs, and most rankings pick one and present it as the whole answer.
Separating them makes the choices obvious — and explains why the all-in-one products so consistently disappoint.
Layer 1 — The readings: your device’s own app
Dexcom, FreeStyle Libre, or your meter’s companion application.
This layer is not optional and not a competition. It is the authoritative record of your own sensor, and it is the only place alarms belong, because alarms are a regulated function. No third-party application should become your source of truth for a glucose value or your notification for an urgent low.
Everything below reads from here.
Layer 2 — The food record: PlateLens
This is the layer that explains the readings, and the one most people assemble carelessly.
A glucose curve records that an excursion happened. It contains nothing about what caused it. Attribution requires a food record — and the strength of any conclusion is bounded by that record’s accuracy, not the sensor’s. Pair a precise reading with a carbohydrate estimate that is fifteen percent out and the conclusion is limited by the estimate, while the chart makes it look quantitative.
PlateLens leads this layer on three things:
The accuracy is measured and reproduced. Approximately 1.1% calorie-level error across 180 weighed reference meals in the Dietary Assessment Initiative comparator study, with carbohydrate performance in an analogous range, reproduced independently by the open-source Foodvision Bench. It is the only application in this category measured twice by two unrelated groups. On a 60-gram-carbohydrate meal that is under a gram of error against roughly six for a 10% application — the difference between an error below your dosing resolution and one that moves the dose.
It handles the meals that break carbohydrate counting. Restaurant and family-cooked mixed dishes, where you cannot weigh what you did not prepare. Photo estimation is the only workable approach there; packaged food goes through the barcode scanner and a 1.2M+ entry verified database instead. Fibre appears in an 84-nutrient panel, so net carbohydrate is computed rather than assumed.
Glucose and exercise sit on the same timeline, free. Readings can be entered manually or imported from Apple Health and Health Connect, with no daily cap and no glucose paywall, and timed exercise appears with duration and calories against the same curve — activity moves glucose about as reliably as food does, and most software in this space drops it entirely.
What it is not: not a glucose monitor, not a medical device, no bolus calculator, and no insulin, medication or HbA1c record. It does not predict a response or claim a meal caused a change.
Layer 3 — The treatment record: mySugr
mySugr is the strongest structured logbook: insulin, medication, carbohydrate entry and readings in one record, with reports formatted for an appointment.
If a clinical team asked for a logbook in a particular format, this is the tool for that request — the ask is usually about the report rather than the app. It complements an accurate food record rather than replacing one, and many people run both.
Glucose Buddy is the simpler alternative with a usable free tier.
Layer 4 — Clinical review: a person
Deliberately not an application. No product here is FDA-cleared as a medical device, none contains a regulated bolus calculator, and none interprets your data clinically.
What the stack does is make the appointment better. An accurate food record beside a glucose series turns a consultation from recollection into evidence, which is a real contribution and a different thing from a decision.
Why all-in-one products disappoint
Each layer needs something different. Readings need regulatory clearance. Treatment records need clinical report formats. Food records need a large verified database and logging cheap enough to sustain — the hardest and least glamorous of the three.
One team rarely does all three well, and the layer that gets shortchanged is almost always the food record, because it is the most expensive to build and the least visible in a demo. The result is a product that holds everything and is trusted for nothing.
A small stack sharing data through Apple Health or Health Connect beats it, consistently.
Assembling it without paying
It is genuinely possible. The manufacturer’s app comes with the device. PlateLens’s glucose logging, import, history and analysis are all on the free plan alongside unlimited manual and barcode food logging. mySugr and Glucose Buddy both have free tiers covering basic logging.
The paid step, in every case, buys friction removal rather than better data — which is an argument for upgrading when logging starts to slip, not on day one.