The algorithm doesn't see you
There is a quiet crisis at the heart of beauty personalisation. Billions of dollars are spent every year on AI-powered recommendation engines, personalised quiz flows, and "smart" beauty platforms — and for the majority of the world's consumers, they simply don't work.
The reason is not complexity. It is data. The models that power most beauty AI were trained primarily on images, clinical trials, and product testing data that skews heavily toward lighter skin tones and European beauty standards. When you feed a recommendation engine data that doesn't represent your skin, the engine can't serve you well — no matter how sophisticated the algorithm.
For consumers with melanin-rich skin — the majority of the world's population — this means receiving recommendations that don't account for hyperpigmentation, uneven skin tone, the specific way darker skin responds to retinoids, or the fact that "SPF 50" products can leave a white cast that makes them unwearable.
The hyperpigmentation blind spot
Hyperpigmentation — the darkening of patches of skin due to excess melanin production — is one of the most common skin concerns globally, and disproportionately affects people with darker skin tones. It can be triggered by acne, sun exposure, hormonal changes, and inflammation.
The products commonly recommended for hyperpigmentation in mainstream beauty circles — high-concentration vitamin C serums, certain forms of retinoids, aggressive chemical exfoliants — can be effective. They can also, if used incorrectly or in formulations not designed for melanin-rich skin, cause post-inflammatory hyperpigmentation: making the very concern they're meant to address significantly worse.
A beauty recommendation engine that doesn't understand this distinction isn't neutral. It's actively harmful.
What melanin-aware personalisation actually means
At Ewamo, we've built our personalisation engine with melanin-rich skin as a primary use case, not an afterthought. In practice, this means several things.
Skin tone as a first-class variable. Ewamo's profile captures not just skin type (oily, dry, combination, sensitive) but skin tone across the full Fitzpatrick scale and beyond — because the same ingredient behaves differently across the spectrum and recommendations need to reflect that.
Ingredient flagging for melanin-rich skin. Certain ingredients that are broadly considered safe have specific cautions for darker skin tones. Our ingredient intelligence layer flags these proactively, with an explanation grounded in the specific mechanism — not just a generic warning.
Formulation awareness. A vitamin C serum is not just a vitamin C serum. Concentration, pH, form of vitamin C, and the rest of the formulation all affect both efficacy and safety for different skin tones. Ewamo's recommendations go to this level of specificity.
African beauty traditions. Shea butter, turmeric, baobab oil, black soap — the African beauty tradition has centuries of effective skincare practice built into it. Ewamo treats these not as exotic additives but as foundational ingredients with genuine efficacy, many of which are specifically well-suited to melanin-rich skin.
Why this matters globally
It's tempting to frame this as a story about Africa or the African diaspora. It is that — but it's larger. The consumers who are underserved by mainstream beauty AI represent the majority of humanity. South Asia, East Asia, Latin America, the Middle East — the world's most populous regions have skin tones that fall outside the implicit standard most beauty AI was built around.
Building a beauty intelligence platform that genuinely serves melanin-rich skin is not a niche play. It is the largest unaddressed opportunity in the global beauty market.
That is what Ewamo is building — starting from a foundation of genuine diversity, rather than trying to retrofit it later.