Ads Are Coming to ChatGPT. Get Your Product Feed Ready.

More of your customers are asking ChatGPT what to buy. Now businesses can pay to be the answer. On 2 June 2026, OpenAI added product feed support to its ChatGPT Ads Manager, a change first reported widely by Search Engine Land and PPC Land on 11 June. Instead of building an ad for one product at a time, a retailer can upload its whole catalogue and have ads generated automatically from each item's name, image, and details.

This is the latest step in a fast rollout. ChatGPT advertising first launched in February 2026, opened up as a self-serve Ads Manager to US businesses on 5 May, and now supports product feeds of up to a million items, with a sample of 100 products required before the full catalogue goes live. OpenAI says feed-based ads have been among the strongest-performing formats in its beta so far.

If you run a small business in Australia, the honest caveat first: you cannot buy these ads yet. The Ads Manager is US-only, with targeting limited to US states, regions, and postcodes, and no international launch date announced. So why should you care today? Because the direction is now unmistakable, and the work that gets you ready costs you nothing to start.

What actually changed

A product feed is just a structured file listing everything you sell: each item's title, description, price, image, availability, and attributes like size or colour. It is the same kind of file that already powers Google Shopping and the dynamic product ads on Facebook and Instagram. OpenAI has added a Feeds section to its Ads Manager so that this file can now drive ads inside ChatGPT too.

The ads appear below ChatGPT's answers, clearly labelled as sponsored, including for people on the free and lower-cost tiers. The idea is to match what you sell to the moment someone is actually deciding what to buy, in the middle of a conversation where they have already described what they want. That is a very different signal from a keyword typed into a search box.

Worth noting: OpenAI tried in-chat checkout first, then shut it down in March 2026 after data from Walmart showed people bought far less when completing the purchase inside the chat than when clicking through to the store. The product feeds built for that checkout were repurposed for advertising instead. The lesson stuck: send people to a shop that is ready for them, do not try to be the shop.

Why this matters from Australia, even now

Two things are true at once. ChatGPT ads are not available here yet, and the way people shop has already changed here. Australians are asking AI assistants for recommendations every day, comparing products, narrowing options, and forming opinions before they ever reach Google or your website. Whether or not you can pay to appear, you want to be the kind of business these tools can understand and recommend.

The practical gift in this news is that the groundwork is identical everywhere. A clean, accurate product feed is now the single asset that feeds Google Shopping, social ads, AI shopping answers, and ChatGPT's new ad format. Tidy it once and you are ready for all of them, including the channels that have not opened in Australia yet.

Your product feed has quietly become your most important marketing asset. It is the one file that decides whether the machines now standing between you and your customer can see what you sell.

What to do now

You do not need a US ad account to get value from this. You need your product data in good shape, because that is what every one of these channels reads. Start here:

• Build or clean up a proper product feed. Every item should have a clear, specific title, an accurate description, a current price, a sharp image, and correct availability. Vague titles like Product A or Blue Item are invisible to the machines doing the recommending.
• Fill in the attributes. Brand, category, size, colour, material, and condition help AI tools and shopping platforms match your products to what someone is actually asking for. Empty fields mean missed matches.
• Add structured data to your website. Product schema markup is the behind-the-scenes label that tells search engines and AI tools the price, stock, and rating of each item. It is free, and it is what lets your own pages be quoted accurately.
• Make your product pages genuinely useful. The checkout lesson holds: the click lands on your site, so the page needs to load fast, answer the obvious questions, and make buying easy. A great ad pointing at a slow page wastes the click.
• Watch how you appear in AI answers. Ask ChatGPT and Google's AI mode the questions your customers ask, and see whether you show up and whether the details are right. That is your early warning system.

None of this depends on a new ad platform opening in Australia. It is the same housekeeping that improves your Google Shopping performance and your AI search visibility today, and it puts you first in line for ChatGPT ads when they do arrive here. The businesses that win the next channel are rarely the ones that spend the most. They are the ones whose data was ready.

This is exactly the kind of work we do at NextAura. We get your product data clean and structured, wire up the feeds and schema that AI tools and shopping platforms rely on, and keep watch as new channels like ChatGPT ads open up, so you are ready to step in rather than scrambling to catch up. Hand us the optimising and the plumbing, and you can get back to looking after your customers.