Clinique
AI face scan shade diagnosis + AR virtual try-on + complementary product recommendation
With Perfect Corp.'s AI face scan and virtual try-on, Clinique reports customers 2.5 times more likely to buy, a basket 30% larger and time on site multiplied by 4 to 5 after try-on.
Key points
- Clinique scans the customer's face to find their foundation shade and try it on in AR.
- In-house color algorithm (Shade Match Science) combined with Perfect Corp.'s AI and AR.
- Conversion x2.5, basket +30% and time on site x4 to x5 after try-on, according to the brand.
- Evidence B (vendor case study); virtual try-on still live on clinique.com in October 2026.
Objective
Remove the main barrier to buying foundation, choosing the right shade, by scanning the customer's face to suggest their shade, then sell more by recommending matching lipsticks and letting customers try on all the makeup without a sample.
The deployment
Clinique, a brand of the Estee Lauder group, built with Perfect Corp. a two-step shade selection journey. The customer has their face scanned; Perfect Corp.'s AI, combined with Clinique's color algorithm, called Shade Match Science, suggests the foundation shade, then recommends three lipsticks matched to that shade. The AR virtual try-on covers the brand's entire makeup range. The setup runs on three surfaces: iPads placed on in-store counters, which play an attract loop and also serve as a conversation opener for beauty advisors; the website, on desktop and mobile; and microsites opened by QR codes placed on displays, badge lanyards and other materials, which let customers continue trying on from home. During the COVID-19 pandemic, Perfect Corp. adjusted its algorithms so that the shade recommendation works even with a mask. According to Jeremy Harris, the brand's head of technology, the deployment covers retail and online markets in Europe, Asia-Pacific, Latin America and North America. The case study is not dated; the virtual try-on page on clinique.com is online as of October 5, 2026.
Results Proof B
Figures from a case study by the vendor Perfect Corp., stated by name by Clinique's VP of technology; the official clinique.com page confirms that the virtual try-on is still in production, without giving any figure.
How it works
Inferred typical approachThe internal detail is not public. Here is a proven approach that leads to the same result, to adapt to your stack.
The stack in detail
- plateforme Perfect Corp. AR Makeup Virtual Try-On Virtual try-on for Clinique's entire makeup range, in store, on the website and on the microsites
- outil Perfect Corp. AI Skin Shade Finder Face scan to determine skin tone, adjusted to work with a mask
- outil Perfect Corp. AI Product Recommendations Recommendation of three lipsticks based on the foundation shade
- outil Clinique Shade Match Science Clinique's proprietary color algorithm combined with Perfect Corp.'s AI
How it runs, concretely
For ops teams-
1Entering the experience customer / beauty advisor
The customer arrives through the counter iPad (attract loop), through the website, or by scanning a QR code on a display or a purchased product.
-
2Face scan and shade selection AI (Perfect Corp. + Shade Match Science)
Perfect Corp.'s AI analyzes the face; Clinique's color algorithm derives the foundation shade from it.
-
3Complementary recommendation AI (product recommendation)
The system suggests three lipsticks matched to the foundation shade found.
-
4AR try-on customer
The customer virtually tries on the foundation, lips and the rest of the makeup before buying.
-
5Maintenance and roadmap Brand Technology team / Perfect Corp.
Clinique and Perfect Corp. evolve the features along a shared roadmap, such as the adaptation to mask wearing during the pandemic.
The match between the detected skin tone and Clinique's shade reference (Shade Match Science). If a shade is poorly calibrated or the capture is distorted (lighting, mask), the recommendation misleads and the conversion gain disappears.
How your customers perceive this type of use
Sourced studiesLe paradoxe est documente des deux cotes : 71% des consommateurs attendent des interactions personnalisees et 76% sont frustres quand elles manquent (McKinsey, 2021), mais 75% declarent ne pas acheter aupres d'organisations auxquelles ils ne confient pas leurs donnees (Cisco, 2024). La « creepy line » est localisee : messages recus quelques secondes apres une recherche et suivi de localisation sont les pratiques qui mettent le plus mal a l'aise (Periscope by McKinsey, 2019).
Acceptance conditions
- La confiance dans le traitement des donnees precede l'achat : 75% ne achetent pas sans elle (Cisco 2024)
- Un cadre legal protecteur rassure : 59% des consommateurs disent que des lois fortes sur la vie privee les rendent plus a l'aise pour partager des informations dans des applications IA (Cisco 2024)
- La personnalisation elle-meme est attendue quand elle est consentie : environ la moitie des consommateurs (US 55%, UK 52%) disent s'inscrire souvent ou parfois a des services personnalises (Periscope by McKinsey 2019)
Red lines
- Le message declenche quelques secondes apres une recherche ou un achat : deuxieme ou troisieme cause de malaise selon les pays (Periscope by McKinsey 2019)
- Le suivi de localisation percu comme de la surveillance : 40% de malaise en Allemagne et au Royaume-Uni (Periscope by McKinsey 2019)
- Le mesusage des donnees personnelles par l'IA, devenu la premiere inquietude des consommateurs, a 53% et en hausse (Qualtrics 2025)
Sources: McKinsey & Company 2021 · Periscope by McKinsey 2019 · Cisco 2024 · Qualtrics 2025
How to replicate
Inference, not sourcedData prerequisites
- Catalog shade reference (foundation, lips) with usable colorimetric values
- Matching rules between foundation shade and complementary products
- Analytics tracking that distinguishes sessions with and without try-on
Org prerequisites
- A technology lead on the brand side who manages the vendor
- Training for in-store beauty advisors on using the tool
- Camera consent and an image non-retention policy validated by the DPO
Possible stack
- Perfect Corp.
- ModiFace
- Banuba
- Revieve
The plan, step by step
- Step 1Start with foundation: calibrate the shade reference with the vendor and test scan accuracy on a panel of varied skin tonesDeliverable: Shade finder validated on foundation
- Step 2Connect the AR try-on to the makeup product pages of the website, desktop and mobileDeliverable: Online try-on with camera consent
- Step 3Add complementary product recommendations based on the shade foundDeliverable: Shade-to-product matching rules in production
- Step 4Measure conversion, basket and time spent by comparing users and non-users, ideally through A/B testingDeliverable: Quantified reading of the effect, corrected for self-selection bias
- Step 5Extend to stores (counter tablets) and link store and home through QR codesDeliverable: Omnichannel journey with QR scan tracking
First step: Launch a shade finder on the best-selling foundation range and measure user conversion against a control group.
The same pattern, proven in other industries
Taobao
Real-time product recommendation with deep learning, served across hundreds of scenarios in the app
Amazon
Item-to-item collaborative filtering product recommendation engine, served across every touchpoint
A.S. Watson (Superdrug, ICI Paris XL, Watsons)
Selfie-based skin diagnosis and product routine recommendation
Sources
- S1 Clinique Boosts Conversions by 2.5 times, 30% Larger Basket Size, and 5x Longer Dwell Time with Perfect Corp. VTO Interested party archive pending
- S2 Makeup Virtual Try On | Clinique Primary archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.