This holiday season, a growing share of online shoppers will skip the keyword searches and filter boxes altogether and head straight for platforms like ChatGPT and Gemini. Instead of clicking through a bunch of blue links and juggling tabs, they’ll fire off a detailed wishlist and let the chatbot do the hunting and comparing for them.
AI-powered shopping is not quite mainstream, but it’s picking up steam with each passing month. In February 2025, 39 percent of the 5,000 consumers surveyed by Adobe said they already used generative AI to shop, and 53 percent said they planned to do so by the end of the year. Adobe also expects AI to drive 517 percent more traffic to retailers’ websites this holiday season compared to 2024. Meanwhile, e-commerce software giant Shopify says that traffic from AI tools to its online stores has grown sevenfold since January, while purchases driven by AI-powered search have jumped elevenfold.
What Is Agentic Shopping?
Agentic shopping, or agentic commerce, is the use of AI agents to act autonomously on a shopper’s behalf, similar to a concierge service. Sometimes used to describe the broader use of AI in online shopping, agentic shopping has started to live up to its moniker with AI agents that can purchase items from third-party websites or trigger a transaction at a specified price point.
In recent months, OpenAI, Google and Amazon have taken things to the next level with agentic commerce, in which AI agents don’t just recommend products but actually act on a shopper’s behalf. Now, people can buy products directly within chatbot platforms and instruct an AI agent to complete the transaction for them. While there are still some issues that need to be worked out with these new features, McKinsey estimates agentic shopping will drive as much as $1 trillion in U.S. retail sales by 2030.
The Rise of Agentic Shopping
From natural language search capabilities to automated checkout, here’s how the biggest players in AI and retail are laying the foundation for agentic commerce.
ChatGPT
In September, OpenAI announced that ChatGPT users in the U.S. could find, compare and purchase products without ever having to leave the platform. The feature, called Instant Checkout, allows customers to purchase one item at a time from Etsy, Shopify merchants and Walmart. The company took a different approach with Target by facilitating purchases within Target’s app on ChatGPT. ChatGPT’s product listings are based on relevance and are not sponsored, according to the company.
When a user comes across an item that is available for Instant Checkout, they must confirm their billing and shipping information before finalizing the purchase. The merchant, who is responsible for shipping the product, pays OpenAI what it describes as a “small fee” for its role in the transaction. OpenAI collaborated with Stripe to develop the Agentic Commerce Protocol that powers Instant Checkout, and it has made the open source tool available to companies who want to participate in Instant Checkout and other AI agents or payment processors.
Google also rolled out new agentic shopping capabilities of its own in November 2025. Shoppers can describe the product they are looking for in AI mode or Gemini, which can search through more than 50 billion product listings in Google’s Shopping Graph. Google’s AI will retrieve relevant product listings with prices, reviews and availability, and its AI agent can call nearby businesses to check if the item is in stock.
Google is also rolling out an agentic checkout feature that lets shoppers track the price of an item and receive a notification if it falls below a certain price. If the product comes from Wayfair, Chewy, Quince and select Shopify stores, the customer can authorize Google to finalize the transaction on their behalf. Similar to ChatGPT’s Instant Checkout, customers can only authorize Google to purchase one item at a time.
Perplexity
In late 2024, Perplexity launched its agentic shopping tool, Buy With Pro, which lets users buy products from select merchants within the platform. The company says product listings are not sponsored, and Perplexity Pro members get free shipping on these purchases.
Perplexity’s AI browser Comet also lets users search for products across the internet and authorize the browser to make purchases on their behalf. Amazon has sued Perplexity for allowing Comet to access its website without permission, saying it degrades the Amazon shopping experience.
Amazon
Amazon also offers agentic shopping features. In February 2024, it launched its Rufus shopping assistant, which can help users find relevant products and answer any questions they might have about those products. In April 2025, Amazon also introduced its Buy for Me feature, which allows shoppers to buy products from third-party websites without leaving the Amazon Shopping app. In addition to suing Perplexity, Amazon reportedly blocks Google’s AI agents from accessing its website.
How Agentic Shopping Could Shape E-commerce
With conversational search, consumers can ask detailed questions and refine their results in a way that’s completely different from traditional search engine practices, RJ Jain, founder and CEO of Price.com, told Built In. This helps consumers find what they’re looking for and discover brands they’ve never heard of before.
AI assistants have the opportunity to become ”powerful new discovery engines” for brands that optimize their product data for AI, Purva Gupta, co-founder and CEO of Lily AI, told Built In. In some cases, smaller brands with well-optimized product listings could outperform legacy competitors. If their product data isn’t optimized, though, brands run the risk of being misinterpreted or generalized by the chatbot.
Even as platforms like ChatGPT aim to capture the entire customer journey, they’re unlikely to replace direct-to-consumer websites. ChatGPT and other AI platforms are more likely to become “a new shelf, not a new store,” Zohar Gilad, co-founder and CEO of Fast Simon, told Built In. In other words, consumers might use these tools instead of Amazon and Google to discover and compare products, but they’ll still go to a brand’s website to make a decision and complete the transaction. Gilad said brands should treat AI platforms as another channel, while designing direct-to-consumer experiences that foster customer loyalty.
Retailers, meanwhile, may be able to reach new customers who are actively shopping for products in a specific category — and therefore more likely to make a purchase. At the same time, though, they risk becoming middlemen who lose control over the design of the shopping experience — and potentially the data that helps them understand their customers’ journey. If shoppers bypass retailers’ websites or apps, those retailers lose the chance to upsell them with add-ons at the checkout. And if retailers hold less sway with consumers, brands might stop advertising with them. While partnering with ChatGPT on its Instant Checkout feature may make sense for companies like Walmart thanks to its massive network of brick-and-mortar stores, Jain said online-only retailers should try to keep transactions on their site, “otherwise it defeats the site completely.”
How Brands and Retailers Can Prepare for Agentic Shopping
Fifty percent of consumers intentionally use AI-powered search, according to one study, and half of those who conduct a traditional Google search receive an unsolicited AI Overview above their results. Be it intentional or not, artificial intelligence is playing an increasingly central role in how people navigate the internet — and that includes online shopping.
As product discovery shifts from search to AI, marketers across all industries have scrambled to pivot from search engine optimization (SEO) to generative engine optimization (GEO): the practice of optimizing web content for the large language models that power AI platforms. Some companies even specialize in helping brands and retailers navigate the GEO frontier. Lily AI, for example, has developed a “product content optimization” platform that makes data “machine-readable and shopper-relevant” to ensure merchants’ products are visible and accurately surfaced across AI platforms.
Unlike the simple search and filter techniques of the past, users’ conversational search queries are analyzed for “context, emotion and intent” on AI platforms, said Gupta, the CEO of Lily AI. Shoppers are typing in hyper-specific queries, like “best running shoes for women over 40 with knee pain,” and merchants need to develop granular product descriptions that help AI models make the connection between the buyer’s needs and the product’s attributes. Without the right metadata, merchants’ products simply won’t show up on these platforms.
Gupta offered these five tips for merchants who want to optimize their product listings for AI platforms:
- Enrich product data beyond product titles and bullet points by including contextual attributes, like style, use case, occasion, mood, materials and benefits.
- Structure your product data for AI by using schema markup, clean taxonomy and connected metadata that supports AI comprehension.
- Blend the structured data and semantic clarity necessary for AI models with the natural language and emotional tone that resonates with humans.
- AI systems reward recency, so refresh product data consistently and continuously.
- Track whether AI systems interpret and present your products accurately.
We’re still in the early days of agentic commerce, so there’s still a lot to be learned about winning visibility and conversion on AI platforms. But if brands and retailers act now, they could potentially gain a first-mover advantage — or risk getting left behind.
“This is not a moment for passive ‘wait and see’ approaches,” Gupta said. “GEO is the new SEO for the AI era and to be visible in chatbots like ChatGPT or Gemini, brands need to speak a language AI understands: structured, accurate, contextual product data.”
Frequently Asked Questions
Which platforms currently offer agentic shopping features?
Several major platforms have launched agentic shopping capabilities, including ChatGPT (with Instant Checkout), Google (through AI mode and Gemini), Perplexity (with Buy With Pro and Comet browser) and Amazon (with Rufus assistant and Buy for Me feature).
What is GEO, and why does it matter for brands?
GEO (generative engine optimization) is the practice of optimizing web content for the large language models that power AI platforms. As product discovery shifts from traditional search to AI, brands need to ensure their product data is structured, accurate and contextual so AI systems can properly interpret and present their products.
