Anyone
Test the beta
Run your own analysis and tell us whether the shade it returns is right. Every level of the scale needs testers, and the deep end needs them most — that is where the industry currently fails and where our claim has to hold.
Take part
We cannot prove
this on our own
faces.
CeySkin has not launched, so there is no community here to show you — no testimonials, no creator features, no follower count. There is something more useful: a specific job that needs people.
Equal accuracy across every complexion is a claim, and a claim is only worth what the testing behind it is worth. That testing needs faces at every level of the scale, particularly the ones the industry keeps getting wrong.
The cohort
Ten levels means ten cohorts.
A test group weighted towards light complexions would return a flattering average and tell us nothing. Every level carries the same weight, and none is optional.
- 01Open
- 02Open
- 03Open
- 04Open
- 05Open
- 06Open
- 07Open
- 08Open
- 09Open
- 10Open
Recruiting at every level
Places are held open across the full scale rather than filled first-come, so the cohort cannot skew light.
Weighted to the deep end
MST 7 through 10 are where existing tools fail and where our claim is most at risk. Those places are prioritised.
Undertones counted too
Warm, cool, neutral and olive within each level — olive especially, since most tools have no category for it.
Four ways in
Useful things you can do now.
Testers
Report a bad match
A mismatch is more useful to us than a compliment. Each report becomes a test case with a measured expected value, so the same failure cannot quietly return in a later release.
Available in the beta
Brands
Publish your shade data
Most shade ranges are described in prose and swatch photographs, which cannot be matched against. Give us measured colour values and full INCI lists, and your products become matchable across every complexion.
Researchers
Check our working
The method is published in full. If our sampling geometry, conversion or thresholds are wrong, we would rather find out from someone who reads this properly than from a user with the wrong foundation.
What you get back
Results, including
the unflattering ones.
Testing that only ever produces good news is marketing. If you give us your time and your face, what comes back is the actual figures — whichever way they fall.
Accuracy by level
Not a single headline figure. Ten of them, one per Monk level, so a strong average cannot hide a weak deep end.
The gap between best and worst
The spread between our strongest and weakest level, stated plainly. This is the number that determines whether the product is doing what it claims.
Known failure cases
The lighting conditions, undertones and edge cases where analysis is still unreliable — documented rather than quietly hoped over.
What changed and why
When a threshold moves or the sampling changes, the reasoning goes out with it.
What we do with your part in it
- Your photographs are deleted within 24 hours, exactly as in the shipping app.
- Nothing you submit is used to train a model.
- Reports are identified by device, never by name.
- You can withdraw and have your data deleted at any point, without asking why.
Take a place on the scale.
Tell us your email and we will write when testing opens for your level. If you are MST 7 to 10, that will be sooner.