Skin homogeneity: what the score actually measures

"Uneven skin tone" shows up in more skincare marketing than almost any other phrase, and it is rarely defined. It is used to describe sun spots, redness, post-acne marks, and general dullness, often in the same paragraph, as if they were one problem with one fix.

Clinically, they are not one problem. Facet's skin module reports a single homogeneity score, but the number underneath it is built from two separate measurements that behave differently and call for different protocols.

Definition

Skin homogeneity is a measure of chromophore variance across the face: how much the concentration of melanin and hemoglobin changes from one point on the skin to another. Chromophores are the light-absorbing compounds that produce skin colour. Melanin drives pigmentation. Oxygenated and deoxygenated hemoglobin drive redness.

• Low variance: skin colour reads as consistent across the face. This is what "even skin tone" describes visually.
• High variance concentrated in melanin: patchiness, sun spots, post-inflammatory hyperpigmentation, melasma.
• High variance concentrated in hemoglobin: redness, rosacea-pattern flushing, broken capillaries, irritation.

A face can have high overall variance from either source, or both, and the two produce the same word ("uneven") on a product label while pointing to opposite protocols.

What actually causes it

The melanin side and the hemoglobin side have different, well-documented drivers.

• Melanin variance: cumulative UV exposure, post-inflammatory hyperpigmentation left behind by acne or injury, hormonal pigmentation such as melasma, and natural age-related mottling.
• Hemoglobin variance: rosacea, reactive or sensitised skin, a compromised barrier, and vascular patterns that become more visible as skin thins with age.

Barrier damage from over-exfoliation, a theme covered in an earlier post on TikTok skincare advice, tends to raise hemoglobin variance specifically, since irritated skin flushes more visibly. It does nothing for melanin variance and can make pigmentation protocols look like they have stalled when the real problem is a different chromophore entirely.

What homogeneity does not measure

• Wrinkles or fine lines. Those are a texture and structure question, not a colour question.
• Pore size. Facet scores pore density as a separate parameter in the same module.
• Acne severity in the moment. Active breakouts are a distinct clinical picture from the pigmentation or redness they leave behind.
• Attractiveness. A homogeneity score is a colour-consistency measurement, not a beauty rating.

How Facet measures it

Facet's skin module computes chromophore variance on the cheek region defined by MediaPipe landmarks, following Fink 2021, the paper behind Facet's skin homogeneity scoring. That research measured chromophore variance across a large sample and found that lower variance correlates with perceived skin health, which is the basis for treating homogeneity as a scorable clinical parameter rather than a marketing phrase. Facet reports the melanin and hemoglobin components separately in the breakdown, alongside pore density and erythema, so a natural or clinical protocol can target whichever chromophore is actually driving the score.

A full example of how the skin module reports this, alongside the other nine modules, is posted at [/sample-report](/sample-report). The citation behind this and every other threshold is listed at [/methodology](/methodology).