Skincare quizzes became common on direct-to-consumer beauty sites around 2019 and 2020. They tried to solve a real problem: people often do not know which products suit their skin, while brands wanted to help without paying for one-on-one consultations. The idea made sense. Its execution has a structural flaw that almost no quiz has addressed.
The flaw is simple: quizzes ask people to describe their skin. Most people are poor judges of what their skin is actually doing.
The self-report gap
Ask, "does your skin feel tight after cleansing?" and the answer measures perception, shaped by the current cleanser, ambient humidity, water temperature, and what "tight" means to that person. Two people with nearly identical barrier function may answer differently because they use different cleansers or prefer different water temperatures.
Ask, "does your face look shiny by midday?" and you capture something associated with oily skin, but also with dehydrated skin producing compensatory sebum, normal skin in humid conditions, or skin over-stripped by a morning routine. The quiz cannot tell these apart. It logs "shiny, yes" and routes the user accordingly.
Recommendations fail at the gap between personal experience and what is happening at the barrier level. Someone with dehydrated skin and midday shine may be sent to oil-control products that strip the barrier further. Someone who calls their skin sensitive after a sting from their last serum may have experienced one-time fragrance irritation, not chronic sensitivity requiring a narrower product range. Self-report cannot separate these cases.
The skin type bucket problem
Quiz systems usually place users in three or four skin type categories: dry, oily, combination, and sometimes normal or sensitive. These labels began as clinical shorthand for practitioners documenting a patient's skin characteristics quickly. They were not created as a product recommendation framework.
Using these buckets for recommendations discards information that matters for product matching. "Combination skin" means someone has both oily and dry zones, but the variation within that label is wide: how oily is the T-zone? How dry are the cheeks? Is the dryness structural or dehydration-related? Is sensitivity present? Is there an active acne component? Categorizing someone as "combination" removes the details that determine which products will work.
A recommendation based on "combination skin" effectively targets the statistical center of that population. For people at the edges of the category, possibly most of those placed there, the result is predictably wrong.
Quizzes are not entirely useless
To be fair, quizzes have uses. For some product categories, especially makeup, matching depends mainly on color and finish preference rather than skin biology, so self-report is reasonable. For broad skincare lifestyle choices, such as time available for a routine, preference for light textures or richer formulas, and favored scent profiles, self-report is the right tool.
The trouble starts when quizzes infer biological skin state and ingredient compatibility from how skin feels and what someone observes. That chain has too many steps where personal perception can diverge from objective skin behavior. Quizzes capture preferences well. They capture biology poorly.
What photo analysis provides that quizzes cannot
A photograph taken in reasonable natural light contains information self-report cannot. It shows actual surface texture in a neutral state, without the filter of personal perception and description. It also shows surface reflectance across facial zones and visible signs of barrier disruption or surface dehydration that someone may miss or blame on another cause.
That is why we built Cutib around photo-based skin reading instead of a longer questionnaire. Image analysis is not infallible, but it presents current skin state more directly and with less perceptual filtering than self-report. A photo shows what skin is doing now. A quiz shows what you think it does, which is less useful for ingredient matching.
The product history layer
Photo analysis asks what your skin is. Product history asks how your skin responds. Both are needed for a complete recommendation.
Using niacinamide in three different products over the past year without problems is a strong sign that you tolerate niacinamide at the concentrations found in consumer skincare. Reactions to two products containing the same specific preservative are also worth flagging, even if neither reaction was severe enough for you to identify the cause.
Quizzes may ask about past reactions, but usually too broadly to extract useful detail. "Have you ever had a reaction to skincare?" produces yes or no. A recommendation system needs the pattern across specific ingredients and product types. That requires rich transaction history or many detailed questions that users may not remember accurately or answer consistently.
What Cutib does not replace
Photo analysis plus product history improves product matching, but does not replace a dermatologist consultation for users with chronic skin conditions, active prescriptions, or a history of severe reactions. With rosacea, significant acne, eczema, or psoriasis, a general product matching tool is only a starting point. You should work with a clinician to understand your skin's needs. The same is true when introducing prescription actives like tretinoin or using professional treatments that change your skin's baseline state. Cutib is for everyday skincare decisions, not clinical skin conditions.
We can narrow product selection to options meaningfully likely to suit your skin, without requiring expertise in cosmetic chemistry or accurate self-assessment of a biological property you have no trained basis to judge. That is a real improvement over a four-option quiz. It is still not a dermatologist visit.