AR virtual try-on for online shopping

Source Sephora (Virtual Artist, with ModiFace) — public case. This is an industry example, not our project
~90%
higher conversion among engaged users
~30%
reduction in makeup returns

The problem

Buying makeup online is high-uncertainty — wrong shades drive returns and dent confidence and conversion.

The AI approach

Virtual Artist uses augmented reality with computer-vision facial mapping and skin-tone analysis to let customers try thousands of shades through their camera, online and via in-store mirrors; later additions include AI skin diagnostics and colour-matching.

Evidence it works

Sephora reports users who engage with the tool are markedly more likely to purchase (some reporting ~3× / “90% higher conversion among users who engage”), a ~30% reduction in makeup returns, and average app sessions rising from ~3 to ~12 minutes; the tool logged 200M+ shades tried on within two years of its 2016 launch.

What “good” looks like

Higher confidence and conversion with fewer returns, and an experience that works consistently across devices and in-store.

Feasibility & cost shape

Typically partner-led (e.g. ModiFace/Perfect Corp); the differentiation is in data quality (pigment/spectral data) and inclusive accuracy, not the AR shell.

Our independent view

Strong on engagement and returns — and strongest when scoped around inclusive accuracy from day one, since that’s both the ethical and the commercial advantage.

Source & attribution

Based on publicly reported information about the Sephora (Virtual Artist, with ModiFace) work.

This is an industry example included for illustration. It is not a Leia Intelligence project, and no client of ours is implied. Figures are as publicly reported by the original parties.

Sources: DigitalDefynd (Sephora case) · Retail Dive · BrandXR · product-insight analysis of try-on accuracy across undertones

This is the class of problem we help organisations tackle — book a call.
Book a call