Online shopping has changed shape roughly once a decade. First ecommerce, then mobile, then social commerce and livestreaming inside apps like TikTok Shop. Each shift moved the starting point of a purchase somewhere new, and each time the brands that noticed early had a few quiet years of advantage before everyone else caught up.
The current shift is agentic AI, and for beauty it may be the most consequential yet. The starting point of a purchase is moving into a conversation with an AI agent, and a conversation is a very different thing to a search results page.
What agentic commerce actually is
An agentic experience is not a chatbot with a better script. A traditional chatbot matches keywords and returns links. An agent uses conversational AI, natural language understanding, and real-time data to answer a shopper's question, recommend products, and carry them through the decision. It can infer the goal behind the question, adapt as the conversation develops, and explain why it chose what it chose.
That last part is what makes it work in skincare. A shopper asking about a serum is rarely asking about a serum. They are asking whether it suits their skin, whether it clashes with what they already use, and whether it will do what the packaging implies. An agent can hold all three at once. A search box cannot.
The adoption data is further along than most teams assume
This is not a forecast about 2030 behavior. It is happening in the current quarter.
In a Bain & Company survey, 44% of US online buyers said they now begin their shopping journey inside a large language model, or split their search between AI tools and traditional search engines. Europe is on the same curve: a McKinsey consumer survey across France, Germany and the UK found 46% using AI to discover or get inspiration for purchases, and 55% using it to learn about a category or product.
Clutch research puts 70% of consumers using AI tools somewhere in the online shopping process, and 65% using it specifically to research products before buying. And in CI&T's Retail Tech Report on agentic commerce, 86% of consumers have either used an AI agent when shopping or are open to doing so, with Gen Z the most willing to try.
Read those together and the conclusion is uncomfortable for anyone still treating this as next year's problem. A large share of the people who will buy from you this season have already asked an AI about the category, and possibly about you.
Why beauty feels this shift before other categories
Beauty has an unusual property: the product is easy to buy and hard to choose. A shopper can be three clicks from checkout and still genuinely not know whether the thing in front of them suits their skin. That hesitation is where the category loses money, and it is precisely the hesitation an agent is built to remove.
It also explains why the personalization numbers are strong here. McKinsey puts the effect of AI-driven personalization at a 15 to 20% improvement in customer satisfaction, and AI-powered shopping assistants at a 20% lift in conversion. We would add our own caveat to any figure like that, including ours: the result depends entirely on the quality of the advice underneath. An agent that recommends confidently and wrongly does more damage than no agent at all.
What this changes for brands
The practical consequence is that product data stops being back-office plumbing and becomes the thing your brand is judged on. When an agent decides what to recommend, it reads what you have published: ingredient lists, suitability, claims, reviews, the answers to the questions people actually ask. Brands with thin, marketing-led product data will be quietly passed over by systems that have no way to be charmed.
The second consequence is about where the conversation happens, and who owns it. A shopper can meet an agent on ChatGPT or Gemini, or they can meet one on your own site. Those are very different commercial positions, and the difference is worth understanding properly before you commit budget to either.
Frequently asked questions
What is agentic commerce?
Agentic commerce is online shopping mediated by an AI agent that can understand intent, recommend products, and guide a shopper through to purchase, rather than simply returning links in response to keywords.
How is an AI agent different from a chatbot?
A chatbot answers a question. An agent holds a goal. It uses natural language understanding and real-time data to interpret what a shopper is trying to achieve, adapts as the conversation develops, recommends specific products, and explains the reasoning behind each choice.
Are consumers really shopping this way yet?
Yes. Bain found 44% of US online buyers starting their shopping journey in an LLM or splitting it between AI tools and search engines, and CI&T found 86% of consumers have used an AI agent when shopping or are open to it.
How big could agentic commerce get?
McKinsey projects that AI agents could facilitate $3 trillion to $5 trillion in global commerce by 2030. Forecasts at that scale are directional rather than precise, but the direction is not in much doubt.
What should a beauty brand do first?
Audit the product data an agent would read: ingredients, suitability, contraindications, and honest answers to the questions your customers actually ask. Everything else depends on that being right.
Read the full piece
This article draws on a longer essay by beauty journalist and trend forecaster Theresa Yee, written for the September 2026 Renude Insights report, which covers agentic commerce in considerably more depth than we can here. You can read the September 2026 Renude Insights report for her full analysis. And if you want to see what an expert-grounded agent looks like on your own site, book a demo.
