Choosing the right AI skin analysis platform means looking beyond the technology behind a selfie. For beauty retailers operating both online and in physical stores, the most important factors are personalised skincare recommendations, product matching across different brands, useful customer insights and seamless integration into the shopping experience.
Dermaself brings these capabilities together by connecting AI-powered skin analysis with cosmetic ingredient information and products from the retailer's own catalogue.
For e-commerce and innovation managers, understanding these capabilities is essential when choosing a technology that can support both customer experience and business decisions.
What is an AI skin analysis platform?
An AI skin analysis platform uses artificial intelligence to evaluate visible skin characteristics and help customers discover skincare products suited to their needs.
These platforms use computer vision, a technology that allows AI to interpret images, to analyse selfies and identify visible skin characteristics such as blemishes, redness, pores and wrinkles.
For beauty retailers, the real value goes beyond identifying skin concerns. It lies in connecting the analysis results with relevant skincare products and information that can support business decisions.
The following five criteria help retailers evaluate which AI skin analysis platform best fits their needs.
1. Personalisation: does the platform match products to individual skin needs?
Effective AI skincare personalisation connects each customer's skin characteristics with the ingredients of products available in the retailer's catalogue.
A personalised experience should help customers understand which skincare products are relevant to their skin, rather than simply identifying visible concerns.
Dermaself combines a skin test, which collects information about the customer's skin, with selfie-based AI analysis. The platform then compares the results with the ingredients of products available in the retailer's catalogue.
This process uses INCI information. INCI stands for International Nomenclature of Cosmetic Ingredients, the standardised names used to list ingredients on cosmetic products.
By connecting skin characteristics with ingredient information, Dermaself recommends formulations suited to individual needs and builds personalised skincare routines.
For beauty e-commerce, this turns large product catalogues into a more guided shopping experience, helping customers discover products relevant to their skin.
2. Brand-independent recommendations: why do they matter for multi-brand retailers?
Brand-independent skincare recommendations suggest products based on skin needs and ingredients, rather than starting from a particular brand.
This approach is especially valuable for multi-brand retailers, businesses that sell products from several beauty brands within the same store or website.
In department stores, for example, customers often encounter dozens of skincare brands and hundreds of formulations. Choosing between them can be difficult without personalised guidance.
Dermaself connects each customer's skin analysis results with suitable products from the retailer's existing assortment, allowing recommendations to include formulations from different brands.
This approach is explored in From selfie to shelf: how AI skin analysis is changing beauty retail, published by Cosmetics Business.
For retailers, ingredient-first matching makes product discovery more relevant to individual customers while keeping recommendations connected to the products they actually sell.
3. Merchandising insights: can AI skin analysis support commercial decisions?
AI skin analysis can help retailers understand customer skin needs and use that information to support merchandising decisions.
Merchandising refers to how retailers choose, organise, present and promote the products they sell.
Beyond individual skincare recommendations, Dermaself provides aggregated data and statistics from customer skin analyses.
Aggregated data means information collected from multiple analyses and combined to identify common characteristics or recurring customer needs, rather than focusing on a single person.
These insights can help e-commerce and merchandising teams:
• Understand customer needs: identify skin characteristics that frequently emerge across analyses
• Inform product assortment: use customer insights when deciding which skincare categories to prioritise
• Improve product presentation: highlight product categories relevant to the needs identified among customers
• Support commercial planning: make decisions using information gathered through skincare consultations
For example, if aggregated analyses reveal that many customers share a particular skin concern, a retailer can consider that information when reviewing its skincare assortment.
This gives Dermaself two complementary roles: helping customers discover suitable products and helping retailers better understand the needs of their audience.
4. Omnichannel integration: can the experience work online and in-store?
An omnichannel AI skin analysis platform makes personalised skincare consultation available across online stores, physical shops and other customer touchpoints.
Omnichannel simply means connecting the different places where customers interact with a brand or retailer, so they can access relevant services throughout their shopping journey.
Dermaself can be integrated into e-commerce websites and connected with the retailer's product catalogue.
In physical retail, customers can access AI skin analysis through QR codes, tablets and smart mirrors. Smart mirrors are digital mirrors equipped with interactive technology that can support personalised shopping experiences.
The analysis can also support beauty advisors, the staff members who help customers choose beauty products, by providing information about visible skin characteristics and relevant skincare recommendations.
For department stores and multi-brand retailers, this creates opportunities to offer personalised guidance across different shopping environments.
The key consideration is how effectively the platform connects skin analysis with product discovery wherever customers interact with the retailer.
5. Scientific validation and privacy: what should retailers verify?
Retailers should choose AI skin analysis platforms supported by documented technical results, clearly explained testing methods and advanced technologies designed to protect customer identity.
When evaluating accuracy, it is important to understand what was tested and how the results were measured.
Dermaself's AI technology has been developed using more than 220,000 annotated skin images. This means that skin characteristics were identified and labelled in the images used to develop the AI models, helping the technology learn to recognise different skin characteristics.
Dermaself internal validation 2026
Dermaself's 2026 internal validation reported two results:
220,000+
annotated skin images
98%
redness assessment within one grade
92%
wrinkle location correctly identified
98% for redness assessment: in 98% of cases, the AI's redness evaluation differed from the reference assessment by no more than one grade. The test involved 121 faces not used to train the model, covering skin phototypes II–IV, categories describing how skin typically responds to sun exposure.
92% for wrinkle location: the AI correctly identified the facial area containing a wrinkle in 92% of cases, across more than 5,000 analysed wrinkles.
Source: Dermaself internal validation, 2026. These results describe specific functions tested under defined conditions.
Image quality also matters. Differences in lighting, smartphone cameras and image settings can affect how skin appears in photographs.
Dermaself's scientific approach is supported by research collaborations with Politecnico di Milano and Politecnico di Torino, exploring advanced image normalisation and privacy-preserving facial analysis.
Image normalisation means adjusting images to reduce differences caused by lighting and camera settings.
Skin Guard: a more advanced approach to facial anonymisation
Protecting customer identity requires more than simply hiding recognisable facial features. Covering elements such as the eyes and mouth may still leave facial information that AI systems can use to recognise a person.
Dermaself's Skin Guard technology takes a different approach. Instead of only concealing individual features, it digitally separates facial skin information from the original face and transfers it onto a neutral, mannequin-like facial model.
This process is designed to preserve the skin information needed for analysis while reducing the identifying information associated with the original face.
For beauty retailers, Skin Guard represents a privacy-focused approach to AI skin analysis, combining personalised skincare technology with an additional layer of identity protection.
Dermaself's research initiatives are also discussed in Cosmetics Business's coverage of Dermaself and Alta Scuola Politecnica.
For retailers, documented performance, responsible image processing and technologies designed to protect customer identity are important factors when evaluating an AI skincare technology partner.
Real-world example: Dermaself at Shaka Beauty Hotel × OVS
During Milan Beauty Week 2026, Dermaself brought AI-powered skin analysis and personalised product recommendations into a multi-brand beauty retail experience.
At Shaka Beauty Hotel, an activation by Shaka, OVS' beauty concept, visitors could complete a selfie-based skin analysis and receive personalised skincare recommendations featuring products from the selected assortment.
The five-day activation recorded
2,632
completed AI skin analyses
75%
completion rate: 75% of users who started the experience completed it
55 sec
one completed analysis every 55 seconds, on average during opening hours
The experience combined physical consultation points with access through visitors' smartphones.
These results highlight strong customer engagement and demonstrate how AI-powered skincare consultation can be successfully introduced into a high-traffic, multi-brand beauty environment.
The activation was also covered by Cosmetics Business in September 2026.
Explore the full Dermaself × Shaka Beauty Hotel case study →
Choosing an AI skin analysis platform: a practical checklist
The right AI skin analysis platform should combine reliable skin assessment, personalised product recommendations, useful customer insights and integration with the retailer's existing shopping experience.
Before selecting a solution, e-commerce and innovation managers should consider six questions:
1. Scientific performance
Does the platform provide documented results showing how accurately it evaluates specific skin characteristics?
2. Personalisation
Are recommendations based on each customer's skin needs and product ingredients?
3. Multi-brand compatibility
Can the platform recommend products across different brands in the retailer's catalogue?
4. Merchandising insights
Can it provide combined customer data that helps retailers understand common skin needs?
5. Integration
Can customers access the experience through e-commerce and physical stores?
6. Privacy
Does the platform use technologies designed to protect customer identity during facial image processing?
These criteria help retailers identify technologies that connect skin analysis with personalised skincare guidance and valuable business information.
Dermaself: connecting skin analysis, product discovery and retail insights
Dermaself combines AI-powered skin analysis, ingredient-first product matching, aggregated customer insights and privacy-focused facial processing to create personalised skincare experiences for beauty retailers.
Its technology connects individual skin needs with relevant products from the retailer's own assortment, while providing information that can support merchandising and commercial decisions.
From e-commerce to department stores and beauty events, Dermaself helps bring personalised skincare consultation into the customer journey.
Discover Dermaself's AI Skin Analysis technology →Frequently Asked Questions
Quick answers to common questions
What is the difference between AI skin analysis and AI skincare recommendations?
AI skin analysis identifies visible skin characteristics using artificial intelligence. AI skincare recommendations use those results to suggest relevant products. Dermaself connects both processes by matching skin needs with the ingredients of products in the retailer's catalogue.
Can AI skin analysis platforms work with multi-brand catalogues?
Yes. Dermaself can recommend skincare products from different brands within a retailer's catalogue by comparing individual skin characteristics with product ingredient information.
How can AI skin analysis improve the department store customer experience?
AI skin analysis helps customers receive personalised skincare guidance while shopping. Through QR codes, tablets and smart mirrors, customers can explore relevant products, while beauty advisors can use analysis results to support consultations.
How does Dermaself's Skin Guard technology protect customer identity?
Skin Guard is designed to protect customer identity by digitally transferring facial skin information onto a neutral, mannequin-like facial model, rather than simply covering features such as the eyes and mouth. This approach aims to preserve useful skin information while reducing identifying facial details.
What merchandising insights can retailers obtain from Dermaself?
Dermaself provides aggregated information from customer skin analyses. This means combined statistics that help retailers identify common skin needs and use those insights to inform product assortment and commercial decisions.
How should retailers evaluate AI skin analysis accuracy?
Retailers should examine the results of specific technical tests, the images used for evaluation and how performance was measured. For example, testing on images excluded from training helps assess how an AI model performs on images it has not previously learned from.