What OpenAI released
OpenAI introduced ChatGPT Shopping Research on November 24, 2025. Instead of returning a short list immediately, the experience can ask clarifying questions, research current product information, and produce a personalized buyer guide.
OpenAI says the system is particularly useful for detail-heavy categories such as electronics, beauty, home and garden, kitchen appliances, and sports and outdoor products. It can use price, availability, reviews, specifications, and images.
Source: Introducing shopping research in ChatGPT.
How the buyer journey changes
Consider:
I need a quiet blender under $250 for nut butter and daily smoothies.
It must fit under a 17-inch cabinet and be easy to clean.
Shopping Research can clarify capacity, region, and preferred features. It then has to compare:
- current price
- dimensions
- motor and use limitations
- container size
- cleaning method
- warranty
- stock and delivery
A product page that says “professional power for every kitchen” supplies almost none of those facts.
Where product information can come from
OpenAI’s help documentation says shopping research may use public product information, other retail sources, and merchant data supplied through the Agentic Commerce Protocol, or ACP.
This means a Shopify merchant should not think only about the webpage. Public product evidence, structured product data, and commerce integrations can all influence what information is available.
Source: Using shopping research in ChatGPT.
What remains outside merchant control
OpenAI states that shopping results are organic, but it also warns that price and availability can be wrong. A merchant cannot control the final shortlist or the wording of the answer.
The merchant can control:
- stable product identity
- complete comparison attributes
- current price and availability
- canonical product URLs
- clear shipping, returns, and warranty terms
- product-feed quality when participating
A practical Shopify response
For the ten products most likely to be compared, create a buyer-fact table. Include dimensions, materials, compatibility, package contents, exclusions, warranty, price, stock, and delivery market.
Store reusable facts in Shopify metafields, display decisive facts in normal page text, and ask the technical owner to compare them with Product JSON-LD and connected feeds.
Then create five unbranded shopping-research prompts. Save the exact prompt, clarifying questions, final products, citations, market, and timestamp. Do not treat one run as a permanent ranking.
For the store workflow, use Which 10 products should you optimize first?.
Community discussion
Add to the article
Ask a technical question, share a storefront result, or challenge a conclusion with evidence.