Start with what actually varies
Masks differ far less on the specifications that get advertised than on the ones that do not. Here is the order we would work in.
1. Documentation quality
This sounds like a strange thing to check first. It is the single most informative signal in the category.
A manufacturer that publishes wavelength peaks, irradiance with a stated measurement distance, session dose, and safety documentation is a manufacturer whose numbers you can compare against another brand’s. One that publishes a large number and nothing else is asking you to take it on trust, and there is no way to check.
You are not evaluating honesty in the abstract. You are evaluating whether the specification you are about to compare means anything.
2. Coverage and fit
The most consistently underrated pair, and the two that most affect whether a device is any use to you.
Coverage is about which parts of your face receive light. Masks vary in how far they extend toward the jawline, whether they reach the perioral area, whether the under-eye region is treated, and whether there is a separate neck attachment. A mask that leaves gaps is not treating those zones at any dose.
Fit determines coverage in practice. A rigid mask sized for a larger face sits further from a smaller one, which reduces delivered irradiance across the whole surface — the specification did not change, but what reaches your skin did. Our guide on how we evaluate facial coverage explains the zone method we use.
If you can try a mask on, or buy from somewhere with a genuine return window, do.
3. Wavelengths, stated specifically
Check that the documentation gives nanometre figures rather than colour names, and that separate wavelengths are listed separately. “Red and infrared” is not a specification; “633 nm and 830 nm” is. See understanding wavelengths.
4. Irradiance, with its measurement distance
If the distance is not stated, treat the number as uncomparable rather than as impressive. That is not pedantry — it is the difference between a figure measured at the LED surface and one measured where your skin actually sits. See how to read irradiance and dose claims.
5. Session length and practical usability
Session length interacts with dose, but it also interacts with whether you will use the thing. A twenty-minute session you must lie still for is a different proposition from a ten-minute session you can walk around during.
Check: is the timer automatic? Can you start and stop without an app? Are the controls usable while wearing the mask? What happens if the app is discontinued in three years?
6. Safety documentation
Look for eye protection guidance, contraindications, and what the manual says about photosensitivity and medications. A device shipping without any of this is telling you something about the manufacturer’s approach.
7. Warranty, returns and support
A twelve-month warranty from a company reachable by email is worth more than a “lifetime” warranty from an anonymous storefront. Check the return window specifically — fit problems only become apparent after wearing the device, which is exactly when a restrictive return policy bites.
8. Price, in that order
Price last, deliberately. Value is what a device delivers relative to cost, and you cannot judge that until you know what it delivers.
Specifications that look useful but mostly are not
LED count. A high count tells you how many emitters there are, not how much light reaches your skin, at what wavelength, or how evenly. Counts also get inflated by counting chips within packages.
Colour count. Seven colours in a fixed area usually means fewer emitters per colour. It is a trade-off presented as a feature.
Total wattage. Power draw, not delivered irradiance.
“FDA cleared” without a document. Clearance is specific to a device and a stated intended use. “FDA registered” or “FDA listed” is an administrative step and is not approval or clearance. See our medical disclaimer on the distinction.
Star ratings on retail listings. Not comparable across products, frequently affected by incentivised reviews, and unrelated to any measurement.
Before-and-after photographs. Lighting, angle, makeup and expression account for more variance than most devices could. We do not publish them as evidence.
Putting it together
The comparison that actually helps looks like this: two or three masks that fit your face, from manufacturers who document what they sell, whose stated wavelengths suit what you are looking for, at prices that reflect what they deliver — and then the one with the better return policy.
Our comparison tables are built to support exactly that, with manufacturer claims and our measurements in separate rows so you can see which brands document what.
Sources
- Citation placeholder: consumer device specification disclosure practices Citation to be added
To be added when supported by our own survey of reviewed devices.