How to get your skincare brand recommended by ChatGPT and Google AI Mode

Why does this matter for a skincare brand right now?

Because the beauty counter has moved into the chat window. NielsenIQ’s State of Beauty 2026 found that 49% of consumers already receive beauty product recommendations from generative AI, and more than half buy through social platforms. A shopper who used to Google “best moisturiser for sensitive skin” now asks ChatGPT, and gets three names back. If yours isn’t one of them, you never had a chance.

The good news: skincare is one of the easiest categories for AI to match. Every product has clear facts: skin type, concern, key ingredients, texture, size and price. Brands that publish those facts clearly get matched. Brands that hide them behind “luxurious, transformative glow” copy don’t.

How do ChatGPT and Google AI Mode decide which brands to recommend?

They read the web, then pick the products that best fit the question. In practice, three things decide it:

1. Fit: can the AI tell your product matches the question? (e.g. “fragrance-free, for rosacea, under $40”)
2. Trust: do other sources agree? Reviews, editorial roundups, Reddit threads, creators and expert mentions.
3. Access: can the AI read your product data? That means product feeds, structured data and pages that aren’t blocked.

Shopping is also changing behind the scenes. In March 2026, OpenAI moved ChatGPT shopping away from in-chat checkout towards product discovery, with shoppers sent to the merchant’s own checkout. Google’s Universal Commerce Protocol powers buying inside AI Mode and Gemini for selected merchants. And Shopify’s Agentic Storefronts let US-eligible stores switch on ChatGPT, Google AI Mode, Perplexity and Copilot from one place.

Step 1: How do I check if AI recommends my brand today?

Spend 30 minutes. Open ChatGPT, Google AI Mode (or Gemini) and Perplexity, and ask the 10 questions your buyers really ask. Use their words, not yours.

For each answer, write down: Were you mentioned? Who was? Which sources did it cite? That list of cited sources is your to-do list.

Step 2: How do I make my product data easy for AI to read?

Put the facts in plain text and in structured fields, not just in images.

• On every product page: skin types, concerns, key ingredients (with % when you can), what it’s free from, size, price, how to use, where it sits in a routine, and when to expect results.
• Structured data: Product schema with offers, ratings and reviews. On Shopify, store skin type and concern in metafields so they feed your theme, filters and feeds.
• Product feeds: keep Google Merchant Center clean and complete. On Shopify, check your Shopify Catalog and Agentic Storefronts settings (US eligibility applies).
• Don’t block AI crawlers: make sure robots.txt allows OAI-SearchBot (ChatGPT search), PerplexityBot and Googlebot.

Step 3: What questions should my own site answer?

The ones shoppers ask AI. Each answer should start with a direct, 1–2 sentence reply, then the detail.

• A short FAQ on every product page (“Can I use this with retinol?”, “Is it safe for pregnancy?” only if you can back it up).
• Ingredient pages: what it does, who it suits, how to use it.
• “Is [product] right for [skin type]?” guides.
• Honest comparison pages, including when a competitor is the better fit.

Step 4: How do I earn the mentions AI trusts?

AI tools lean on what others say about you. Since only about 38% of AI Overview citations come from top-10 Google results, being mentioned in the right places matters as much as ranking.

• Reviews: volume, recency and detail (skin type, results, time used). Ask for them in your post-purchase emails.
• Editorial roundups: “best moisturisers for sensitive skin” lists from beauty publishers.
• Communities: real, disclosed answers on Reddit and skincare forums. Never fake accounts.
• Experts and creators: dermatologist and creator content that names your product and why.

Step 5: How do I keep AI from repeating claims that get me in trouble?

AI repeats what you publish. So publish only what you can prove. Keep cosmetic claims cosmetic (“reduces the look of dark spots”), avoid drug claims (“treats acne”, “cures eczema”) unless your product is regulated for it, and never post fake or incentivised reviews without disclosure. In the EU, AI-generated people in ads must now be clearly labelled (EU AI Act, from 2 August 2026).

Step 6: How do I measure it?

• Repeat your 10-question check every month and track how often you’re named.
• In GA4, watch referral traffic from chatgpt.com, perplexity.ai and gemini.google.com.
• Add “AI assistant (ChatGPT etc.)” to your “How did you hear about us?” question.

The numbers

How we’d approach this at Cliktron

• Plan: run the AI visibility check and pick the 10 questions that matter most for your best-sellers.
• Convert: rebuild product pages with clear facts, FAQs and structured data (Shopify & CRO).
• Attract: line up reviews, creators and roundups around those questions (Creative & UGC).
• Retain: ask for detailed reviews in post-purchase emails (Email & SMS).
• Measure: track mentions and AI referral traffic every month (Tracking & Analytics).

Your next step

Common mistakes

• Writing mood copy instead of facts. AI can’t match “dewy glow ritual” to a question.
• Putting key info only in images.
• Blocking AI crawlers by accident.
• Buying fake reviews. It’s illegal under the FTC’s review rule and the UK’s DMCC Act, and it can get you removed.
• Checking once and stopping. AI answers change every month.