What’s New in ChatGPT’s Shopping Feature
Users can now use OpenAI’s new shopping feature to ask ChatGPT for product recommendations in the home decor, beauty, fashion, and electronics industries. You can search for the “best wireless headphones under $100” or “summer sandals for flat feet” and get curated lists with images, reviews, prices, and direct links.
And, most importantly, it’s not only for paid users. This is being rolled out across Free, Plus, and Pro accounts, and it works even when not logged in. That’s a massive funnel, arguably wider than many shopping platforms right now.
How It Works
The recommendations come from structured metadata pulled from across the web. Think of it like Google Shopping but contextual and stripped of ads. Rather than being heavily focused on keywords, ecommerce on ChatGPT is more conversational, focused on how people are searching for these products, and personalized, remembering your previous searches to help provide recommendations for your preferences.
Why This Matters for Marketers and Merchants
This isn’t just another “AI does X” update. It’s a shift in how products are discovered: when, where, and why they show up.
Traditional platforms like Google, Meta, and Amazon have built incredibly effective systems around ads, targeting, and marketplace visibility. That works, and it’s probably not going anywhere anytime soon. However, there is a different flow to sell products on ChatGPT. Right now, product results aren’t driven by ad spend. They’re shaped by structure, clarity, and how well your product information, reviews, and content hold up in a conversation.
That makes this less about performance marketing and more about product content and trust signals. If the data is vague, the reviews are stale, or the schema is incomplete, you may not get surfaced. And that matters, because some users won’t switch back to traditional search; they’ll stay in the conversation, where it’s easier to ask, refine, and decide.
For ecommerce brands, especially direct-to-consumer (DTC) and niche industries, this could be a powerful new discovery. There’s still time to stand out before competition and potential monetization models tighten things up, so don’t hesitate to work with a top-rated SEO company to help ensure you have checked all the boxes you need to succeed.
What You Should Do Next
Here’s what’s worth doing right now to sell products on ChatGPT before everyone else catches up.
Clean Up Your Product Data
AI tools like ChatGPT don’t magically understand your store. They rely on structured data like prices, images, stock levels, and ratings to know what you’re selling. If your schema is messy or missing entirely, you won’t get surfaced. Tools like Schema.org or Google’s Rich Results Test can be used to see what you’re actually feeding the machines.
Get Your Reviews in Order
More is not better. ChatGPT is pulling context, so reviews that are older, vague, or look fake will do more harm than good. Fresh, detailed, authentic feedback helps train AI to trust your product.
Sync Your Catalog
OpenAI’s clearly moving toward in-chat checkout, and Shopify looks like it might be first in line. Developers recently spotted a plugin in ChatGPT’s backend pointing to Shopify’s checkout. It’s not a matter of if. It’s a matter of when. So your product listings need to be accurate, up to date, and ready to be pulled in at any moment. Don’t wait for a big integration announcement. You’ll already be playing catch-up.
Put Your Prompts Through ChatGPT to See If You Appear
Yes, really. Treat it like rank tracking. Ask ChatGPT for recommendations in your category and see who comes up. Those are some of your new competitors.
Rethink How SEO Fits Into This
This isn’t about ranking for a keyword on Google anymore. It’s about being visible to AI which acts more like a concierge. Invest in structured content and helpful product information that plays well in conversational queries.
How This Impacts Attribution and Search Strategy
Let’s zoom out. If ChatGPT starts delivering real ecommerce results, and users stick around to buy, you’re going to lose visibility on how people found you. That may be a problem for anyone running paid media or managing ROAS expectations.
Expect analytics gaps. Referral data might not reflect ChatGPT traffic. Cross-channel performance might feel “off” if conversions spike but no one can trace them back. You’ll need to start triangulating based on behavior and adjusting your reporting logic.
And if you’re still treating SEO like blog posts and meta tags, you’re 10 steps behind. This is structured ChatGPT ecommerce content now. Think: product page quality, feed health, third-party trust indicators, and conversational readability. Less keyword stuffing. More substance.