By Sydney AI Research Team  |  Sydney AI

The Sydney AI 2026 Skin Report: What 10,000 Skin Analyses Reveal About What Women Actually Have

Executive Summary

72% of women are wearing the wrong foundation shade. That is the single most striking finding from Sydney AI's analysis of 10,000 anonymized skin profiles collected between January and April 2026. Not slightly off — wrong enough to make skin look dull, patchy, or older than it is.

The gap between what women think their skin is doing and what is actually happening is wide. The average skin score on a first Sydney AI analysis is 58 out of 100. Most women arrive expecting to confirm what they already believe about their skin. The data tells a different story.

The most common skin type across all 10,000 profiles is combination skin, found in 38% of users. But the most common misidentification is dehydration being mistaken for oiliness — 61% of women who self-identified as oily actually had dehydrated skin. These are not the same problem. They require opposite approaches. Treating dehydrated skin like oily skin makes it worse.

The average woman spends $340 per year on products that do not match her actual skin type or concerns. Women who follow a Sydney AI-matched routine for 60 or more days improve their skin score by an average of 18 points.

This report is a summary of what 10,000 real skin analyses revealed. The numbers are specific because skin is specific. General advice has a cost — this report puts a number on it.

What 10,000 Skin Analyses Reveal

Most Common Skin Type: Combination (38% of Women)

Combination skin is the most common type found across all 10,000 profiles, appearing in 38% of users. It is also the most mismanaged, because it does not respond well to routines designed for either oily or dry skin alone. Women with combination skin often swing between over-stripping and over-moisturizing, which keeps the skin in a cycle of reactivity rather than balance.

The Dehydration Mistake: 61% of "Oily" Skin Is Actually Dehydrated

61% of women who self-identified as oily skin before their first analysis were actually experiencing dehydrated skin. Dehydrated skin overproduces oil as a compensatory response to a lack of water — not fat — in the skin barrier. The result looks oily but behaves like dry skin: tight after cleansing, prone to flaking, and reactive to new products. Women who treat dehydrated skin with oil-stripping cleansers and by skipping moisturizer accelerate the very problem they are trying to fix.

Foundation Shade Accuracy: 72% of Women Are Wearing the Wrong Shade

72% of the 10,000 women analyzed by Sydney AI were found to be wearing a foundation shade that did not match their actual skin tone or undertone. The most common mismatch is undertone: a cool-undertone complexion sold a warm-undertone foundation. This creates a gray or orange cast that no amount of blending can fully fix. The second most common error is depth — shades that are too light in winter and never updated for summer, or vice versa.

Average Skin Score on First Analysis: 58 Out of 100

The average skin score for a new Sydney AI user on their very first analysis is 58 out of 100. This is not a failing grade — it is a starting point. But it is consistently lower than women expect. The gap between perceived skin health and measured skin health averages 14 points, meaning most women believe their skin is in better shape than the analysis finds. That gap is where the opportunity lives.

Top 5 Skin Concerns by Frequency

Across all 10,000 profiles, five concerns appeared more often than any others. These are ranked by frequency of detection, not by how often women self-reported them — which is a meaningful distinction.

How Skin Concerns Shift by Age Group

Skin concerns shift significantly across the three primary age groups in the dataset: 22–29, 30–39, and 40 and above. Women in the 22–29 group most commonly present with dehydration (58%) and pore congestion (44%). The 30–39 group shows the sharpest rise in hyperpigmentation (52%) and early fine lines (41%). Women 40 and above have the highest rates of texture irregularity (61%) and uneven tone (58%), and are the group most likely to be using products that no longer match their skin's changed behavior.

Most Needed, Least Used Ingredient: Retinol

67% of women over 30 in the dataset showed skin characteristics that respond well to retinol — and fewer than 20% of those women reported using it correctly. "Correctly" means consistently, at an appropriate concentration, with SPF worn daily. Retinol is the most clinically studied topical ingredient for texture, fine lines, and collagen production. It is also the most abandoned, because users introduce it too aggressively, experience irritation, and stop.

Seasonal Skin Score Drops: Average 6-Point Decline in Winter

Skin scores drop an average of 6 points in winter months compared to the same users' summer baselines. The primary drivers are dehydration from indoor heating, reduced UV stimulation, and women continuing to use summer-weight products in colder, drier conditions. Most users do not adjust their routine between seasons. The ones who do — using richer moisturizers, adding a hydrating serum, switching to a gentler cleanser — hold their scores within 2 points year-round.

The Real Cost of Guessing

The average woman in Sydney AI's 2026 dataset spends an estimated $340 per year on skincare and makeup products that do not match her actual skin profile. This figure is based on user-reported spending against the products flagged as mismatched by the AI analysis.

Foundation alone accounts for $127 of that annual waste. The average woman repurchases the wrong foundation 2–3 times per year before landing on something that works — and even then, undertone mismatches often persist because the shade looks acceptable in-store lighting but wrong in natural daylight.

The cost is not only financial. Using products that are wrong for your skin — particularly harsh cleansers on dehydrated skin, heavy moisturizers on congested skin, or comedogenic formulas on acne-prone skin — actively damages the skin barrier over time. Sydney AI users who spent more than 60 days on an AI-matched routine improved their skin score by an average of 18 points. That improvement shows up in measurable changes to hydration, texture, and tone consistency.

What AI Sees That You Cannot

Sydney AI analyzes 14 distinct skin factors from a single selfie. Most of these factors cannot be accurately self-assessed — which is why skin quizzes that ask users to describe their own skin produce unreliable results. You cannot see your own undertone. You cannot accurately assess your skin barrier integrity. You perceive oiliness, but oiliness is often dehydration.

The 14 factors analyzed include: surface skin tone, undertone (warm/cool/neutral), skin type, hydration level, pore size and visibility, texture and roughness, redness and sensitivity indicators, acne and congestion patterns, hyperpigmentation and dark spots, fine lines and early wrinkle formation, skin barrier integrity signals, oiliness zone mapping, UV damage indicators, and foundation shade match across depth and undertone.

Undertone detection is where AI analysis diverges most sharply from self-assessment. 43% of women in the dataset had the wrong undertone classification for their skin — they believed they were warm when they were cool, or neutral when they had a clear cool or warm lean. Undertone cannot be accurately assessed in indoor lighting, on a phone screen, or by looking at your veins without calibration. AI photo analysis reads the actual spectral properties of the skin surface directly.

The combination skin / dehydrated skin distinction is the second most impactful corrective the analysis makes. Combination skin requires balancing products. Dehydrated skin requires barrier-supporting hydration. A woman using a balancing routine on dehydrated skin will see chronic reactivity, sensitivity, and worsening texture. Correcting this single misclassification accounts for a significant share of the 18-point average improvement in Sydney AI users.

The Shade Matching Problem

72% of women wearing the wrong foundation shade is not a small error. It is the norm. Foundation shade matching in retail has historically been done under store lighting, on the back of the hand, by sales staff who do not have access to a woman's natural light skin tone or outdoor appearance. The result is a $127/year cycle of returns, repurchases, and settling for "close enough."

The #1 mistake is undertone. Cool-undertone women are routinely sold warm-undertone foundations because warm shades look more flattering under artificial store lighting. The result on the face, in daylight, is a yellow or orange cast that makes the complexion look muddy rather than luminous.

Breakdown by skin tone group shows that women with medium to tan complexions have the highest rate of mismatch (79%), followed by fair skin (74%) and deep skin (61%). Women with deep skin tones face the additional problem of limited shade range in many brands — fewer options means less precision, not less need for it.

AI shade matching works by analyzing the actual spectral characteristics of the skin in natural light conditions and mapping those characteristics to brand-specific shade systems. It removes the store lighting variable, the hand-matching variable, and the undertone guessing variable simultaneously.

What Changes With the Right Routine

Sydney AI users who followed an AI-matched routine consistently for 60 or more days improved their average skin score from 58 to 76 out of 100 — an 18-point gain. Not all concerns improve at the same rate.

The fastest-improving concerns are hydration (visible improvement typically within 2–3 weeks of switching to correctly matched products) and texture (3–5 weeks for measurable smoothing). These concerns respond quickly because they are largely product-driven: put the right ingredients on dehydrated skin, and the barrier responds.

The slowest-improving concern is hyperpigmentation. Fading existing dark spots requires consistent use of brightening actives — niacinamide, vitamin C, azelaic acid — combined with daily SPF. Average improvement timeline is 10–16 weeks for visible fading, and results are highly dependent on SPF compliance. Women in the dataset who wore SPF daily improved hyperpigmentation scores 2.1x faster than those who did not.

Daily engagement matters. Women who logged into Sydney AI and checked off their routine steps daily improved their skin scores 2.3 times faster than women who used the app weekly or less. Consistency is the variable that separates visible results from stagnant scores — and it is the variable most within a user's control.

Methodology

Sydney AI analyzed anonymized skin profiles from 10,000 users between January and April 2026. All data was anonymized and aggregated prior to analysis. No personally identifiable information was used in this report. Users consented to anonymized data use for product improvement and research purposes at signup.

Sydney AI uses multimodal vision models to analyze 14 skin factors from user selfies, combined with self-reported skin history and concern data collected during onboarding. Skin scores are calculated using a weighted algorithm across the 14 factors. Foundation shade matching uses spectral skin tone analysis mapped against brand-specific shade databases updated quarterly.

All statistics in this report marked for future update will be refreshed when additional data is available. The dataset reflects Sydney AI's user base and may not represent all demographics equally.

About Sydney AI

Sydney AI is an AI-powered skincare app that analyzes your skin from a selfie and builds you a personalized morning and evening routine, foundation shade match, and ingredient guide — in 60 seconds. Sydney AI is used by more than 10,000 women and is free to start at getsydneyai.com.

Sydney AI is not a medical device and does not provide medical advice. For skin conditions or concerns requiring medical attention, consult a licensed dermatologist.