Estée Lauder virtual services conversion
Estée Lauder Virtual Services Conversion: What It Means for Virtual Makeup Try-On
Estée Lauder reported that its virtual services convert at up to 10 times the average. It is their number, not ours — but the mechanism behind it explains why photo-based virtual try-on belongs on your product page, quiz and WhatsApp.
Few public numbers in beauty tech are as often quoted, and as often misquoted, as the one The Estée Lauder Companies gave on an earnings call in late 2020. It is worth reading it carefully, because what it says about Estée Lauder virtual services conversion is more useful to a beauty brand than the headline suggests — and it points directly at why virtual makeup try-on earns its place in a funnel.
What Estée Lauder actually reported
On the company's first quarter fiscal 2021 earnings call, CEO Fabrizio Freda described the performance of the group's virtual services — virtual try-on and online consultations — saying they have a conversion rate of up to 10 times the average and a higher average order value, and that in the first fiscal quarter of 2021 alone there were more than 1 million virtual try-on sessions, with more than 30 minutes per session (Fabrizio Freda, CEO of The Estée Lauder Companies, Q1 fiscal 2021 earnings call, November 2, 2020).
Two caveats, stated plainly.
First, this is a third-party number. It is Estée Lauder's result, reported by Estée Lauder, for Estée Lauder's brands, catalog and traffic. It is not a MaIA result, and no vendor — including us — can promise you that outcome.
Second, "virtual services" is broader than try-on. It includes online consultations with advisors. Narrowing the quote to "virtual try-on converts 10x" misrepresents what was said. The honest reading is that a family of guided, personalised digital experiences converted far above the site average and produced long, engaged sessions.
That broader reading is actually the more interesting one, because it tells you what to build.
Why the mechanism travels beyond one company
Strip the number away and look at the behaviour underneath it. A shopper on a lipstick or foundation page is stuck on one question: is this shade for me? Until that question is answered, the add-to-cart button is a gamble. The shopper either abandons, or buys and returns, or opens six tabs to compare swatches on skin tones that are not theirs.
Virtual services — whether a live consultation or a try-on — remove that question from the path to purchase. They replace an abstract hex swatch with an image of the shopper's own face. The long session times in the Estée Lauder quote (Fabrizio Freda, CEO of The Estée Lauder Companies, Q1 fiscal 2021 earnings call, November 2, 2020) are a clue: people were not bouncing, they were deciding.
That mechanism does not require a specific technology. It requires that the shopper sees the product on themselves, on a shade range calibrated to the brand's real SKUs, at the moment of doubt. Photo-based try-on does exactly that.
Where photo-based virtual try-on removes the doubt
MaIA Try-On has two forms, and they fit different moments.
The API is photo-based: you send a photo and the product — lipstick or foundation — comes back rendered on it, using a palette calibrated to your real products in your project. Nothing about it requires an app install, a device certification programme or a camera permission prompt.
The browser SDK, in early access since September 2026 and granted per account, runs in the shopper's browser. It covers thirteen products — lipstick, foundation, blush, face powder, under-eye concealer, eyeshadow, mascara, eyebrow makeup, eyeliner, bronzer, contour, highlighter, lip liner — plus a combined Full Look, in live camera mode or photo mode, with any #RRGGBB colour, self-hosted by the brand. Those eleven extra products and the Full Look are SDK-only local previews; the API does not render them. If you are weighing the two approaches, we compared them in detail in photo-based virtual try-on vs AR try-on.
On the product page
The doubt is at its sharpest here. A photo upload or a saved selfie turns a shade grid into a decision. For brands that cannot ship a front-end component quickly, the photo-based API is the lighter integration: your page already handles the upload, you just call the endpoint and show the returned image.
In the quiz
A foundation quiz that ends in a shade name ends in doubt. A quiz that ends in the shopper's own face wearing that shade ends in a decision. Pair try-on with a beauty profile from the MaIA Scan API — skin tone and undertone, skin type and condition, hair attributes, each with a confidence score, trained on Brazilian skin — and the quiz recommends a shade instead of asking the shopper to self-diagnose it.
On WhatsApp
WhatsApp is where the "virtual consultation" half of the Estée Lauder quote maps most naturally in Brazil. The shopper sends a selfie, receives the shade rendered on their face, and asks a follow-up question in the same thread. MaIA Beauty Advisor covers the consultative layer — an AI advisor integrated with your catalog that reads the beauty profile and recommends your own products — delivered as a project with the B4A team rather than as a self-serve API.
One boundary worth stating to your legal and content teams: MaIA describes visible attributes in a photo. It does not diagnose skin conditions, treat anything, or replace a dermatologist or a professional advisor.
What you should actually measure
Do not import someone else's conversion multiple into your business case. Measure your own:
- Try-on engagement rate — sessions that open the experience, by page and by channel.
- Conversion of try-on users vs non-users, segmented by traffic source, so you separate the tool's effect from the fact that high-intent shoppers use tools.
- Average order value for both groups.
- Return rate on shade-sensitive SKUs, before and after.
- Shade distribution — whether try-on is pushing sales toward the shades your shoppers actually match, or just toward the hero shade.
Conversion depends on your catalog, your traffic and your channel. There is no universal multiple. What there is, is a mechanism with a clear logic and a way to measure it inside your own funnel — which is exactly what we set up during a project.
Ready to test it on your own shades? Create an account and an API key in the MaIA API documentation, or talk to our team about a project across product page, quiz and WhatsApp.
Frequently asked questions
What exactly did Estée Lauder report about virtual services conversion?
On the Q1 fiscal 2021 earnings call, CEO Fabrizio Freda said the group's virtual services — virtual try-on and online consultations — have a conversion rate of up to 10 times the average and a higher average order value, with more than 1 million virtual try-on sessions in that quarter alone and more than 30 minutes per session ([Fabrizio Freda, CEO of The Estée Lauder Companies, Q1 fiscal 2021 earnings call, November 2, 2020](https://www.fool.com/earnings/call-transcripts/2020/11/02/the-estee-lauder-companies-el-q1-2021-earnings-cal/)). This is their reported number, for their brands and traffic.
Can MaIA Try-On deliver the same conversion result?
No one can promise that. The figure above is a third-party result reported by The Estée Lauder Companies, not a MaIA result. Conversion depends on your catalog, your traffic and the channel, which is why we define and measure the baseline inside each project.
Is MaIA Try-On augmented reality?
No. The MaIA Try-On API is photo-based: you send a photo and receive the product rendered on it. The browser SDK adds a live camera mode rendered locally in the shopper's browser, with nothing uploaded.
Which products can be rendered?
The API renders lipstick and foundation on a photo, with palettes calibrated to your real products. The browser SDK, in early access since September 2026 and granted per account, covers thirteen products plus a combined Full Look as local previews in the browser.
Do I need MaIA Scan to run a try-on?
Not necessarily, but they work well together. MaIA Scan returns a beauty profile — skin tone and undertone, skin type and condition, hair attributes, with a confidence per field — so a quiz can recommend a shade instead of asking the shopper to guess it, and then show it on their face.