Beauty brands need six pieces of technology to deliver personalized skin analysis and product recommendations: photo capture, a skin analysis engine, a short quiz, product matching rules on the catalog, a place to show results, and links to email, CRM and analytics. Most brands should buy the analysis and own the rest. Face Age covers all six as one install on Shopify or any website.
This question shows up in our search data almost word for word, and behind it is usually a simpler one: "what do we actually need to buy?" The honest answer is less than vendors suggest, but more than a camera and an AI model. Here's how the pieces fit.
Key terms
- Skin analysis engine: the software that reads a face photo and returns visible concerns with severity levels.
- Matching rules: the links between your products and the concerns they target, plus rules that include or exclude products.
- Zero-party data: information a shopper chooses to tell you, such as quiz answers about skin type or goals.
- Result page: the screen a shopper sees after the scan, where products, routines and reports appear.
- Stack: the set of tools a brand uses together, such as its store platform, email tool and analytics.
The six layers at a glance
| Layer | What it does | Build or buy? | In Face Age |
|---|---|---|---|
| 1. Photo capture | Gets a usable selfie on any phone | Buy | Camera step in the scan, on desktop and mobile |
| 2. Analysis engine | Reads visible concerns and severity | Buy | 13 concerns across 250+ attributes, Skin Age, Skin Health |
| 3. Quiz | Asks what a photo can't show | Buy the tool, write the questions yourself | Quiz builder, before or after the scan |
| 4. Matching rules | Connects results to your products | Own the decisions | Concern and severity mapping, include and exclude rules |
| 5. Delivery | Shows results where shoppers buy | Buy, then customize | Result page, product page blocks, routines, PDF |
| 6. Data and follow-up | Sends results to your tools | Connect what you have | Klaviyo, HubSpot, Mailchimp, Shopify Customers, analytics, webhooks |
1. Photo capture
Everything downstream depends on the photo. Capture has to work on cheap phones, in bad bathroom light, for people who've never done this before. Look for clear framing guidance and a check that rejects unusable photos instead of analyzing them anyway.
This isn't where a brand should spend engineering time. It's solved, and getting it wrong is expensive in support emails.
2. The skin analysis engine
This is the part people mean when they say "AI skin analysis." It reads the photo and returns visible concerns, each with a severity level. Face Age reads 13 concerns, including fine and deep wrinkles, dark circles, pores, acne, pigmentation, redness and oiliness, and adds Skin Age and Skin Health. The full list is on the AI skin analysis page.
Two things matter more than the headline number of concerns. First, consistency: the same face in the same light should get close results twice in a row. Second, honest framing: the output should guide shopping, not sound like a medical opinion. Face Age is cosmetic. Not a medical diagnosis; does not replace a dermatologist.
Building your own engine means collecting and labeling a large set of face images across ages and skin tones, then keeping it up to date. For a brand that sells skincare, that's rarely a good use of money.
3. The quiz
A camera can't tell you that a shopper reacts to fragrance, is pregnant, wants a three-step routine or has a budget. A short quiz can. The best quizzes are short and only ask questions that change what you'd recommend.
Quiz answers are also zero-party data, which is useful well beyond the result page. Our guide to zero-party data from a skin scan and quiz covers what to collect and how to use it. Face Age's quiz builder runs before or after the scan.
4. Matching rules on your catalog
This is the layer you should own, because it's where your product knowledge lives. Each product is linked to the concerns it targets and the severity it suits. Quiz answers then include or exclude products. In Face Age, exclusions always win, so a "sensitive skin" answer can keep a strong exfoliant off the list whatever the scan found.
You can map products by hand or let AI suggest a first pass, then review it. We wrote a step-by-step guide on connecting AI skin analysis to your product catalog that walks through the spreadsheet, mapping and rules.
5. Delivery: where the results show up
The result page is the obvious spot: matched products with concern tabs and "Add all to cart". But personalization works better when it follows the shopper:
- a Product Insight block on product pages that explains why an item matches their scan,
- a carousel of recommended products,
- AM and PM routines built from your catalog, on Plus and above,
- a branded PDF report they can download, on Plus and above,
- a discount code shown right after the result, if you use one.
Make sure you can edit the wording everywhere. You'll want your brand voice, and you'll want to remove anything that sounds like medical advice.
6. Data and follow-up
A scan that disappears when the tab closes is a missed chance. Two things help here.
History and comparison. On Plus and above, Face Age keeps analysis history in the browser, on the server or both, so returning shoppers can compare two scans side by side.
Connections to your stack. Results, leads and events can go to Shopify Customers, Klaviyo, HubSpot, Mailchimp, Omnisend, Google Analytics, the Facebook Pixel and webhooks. That lets your email flows and CRM use the shopper's top concern, not just their last purchase. On Shopify, orders placed after an analysis are tagged, and the analytics page shows which matched products get visited and added to cart.
Build, buy, or something in between?
There are three common routes:
- Buy an app or plugin. Fastest. You install, map products and write the quiz. Face Age works this way on Shopify, WooCommerce and WordPress.
- Use an API or SDK in your own front end. More control over design, more developer time. Our post on skin analysis API vs Shopify app explains the trade-off when the features are the same.
- Build the whole thing. Only makes sense if skin analysis is your core product, not a way to sell your skincare.
Whichever route you choose, the work you can't hand off is the same: deciding which product is right for which concern, writing good questions and keeping the wording honest.
Questions to settle before you pick a vendor
- Which store platform are you on now, and which will you be on in two years?
- Who on your team will own the product mapping?
- Where should scan history live, and what does your privacy policy say about it?
- Which tools must receive scan results on day one?
- Do you need history and reports at launch, or only the scan and products?
For a fuller checklist, see how to choose an AI skin analysis tool for your beauty store. If you're comparing named vendors, our 2026 guide to AI skin analysis tools for ecommerce stores lays them out side by side.
FAQ
What technology do beauty brands need for personalized skin analysis? Photo capture, an analysis engine, a quiz, matching rules on your catalog, a place to show results, and links to your email, CRM and analytics tools.
Should a beauty brand build its own skin analysis? Usually not. Buy the analysis and own the mapping, quiz, wording and follow-up.
Is a quiz enough without a skin scan? It works, but it relies on self-description. A scan adds what the camera can see.
Does Face Age need a custom build? No. It installs as a Shopify app, a WooCommerce or WordPress plugin, or a script on any site.
Is AI skin analysis a medical tool? No. Not a medical diagnosis; does not replace a dermatologist.



