We analyzed 1.18 million prompts and a 7,500-prompt labeled sample to understand exactly what makes ChatGPT surface Shopping cards — and the answer surprised us.
Category beats intent
Our central finding: ChatGPT Shopping is driven far more by product category than by purchase intent language. Prompts naming shippable consumer goods — electronics, apparel, home goods, sports equipment — were dramatically more likely to trigger Shopping cards, while software, services, travel, and financial products almost never did, regardless of how purchase-ready the language sounded.
This is a meaningful departure from how traditional e-commerce advertising works. A prompt like "I'm ready to buy a CRM for my team" produced zero Shopping cards. A prompt like "wireless earbuds under $100" produced them reliably.
Constraints increase probability — but only in the right categories
Adding 1–2 concrete constraints (price range, specific feature, use case) increased Shopping trigger probability, but only within product categories that were already eligible. Constraints alone can't unlock Shopping for service-category queries.
The simple rule that explains 95% of behavior
If the main noun in a prompt is something you could buy on Amazon and have shipped to your house, Shopping is likely to appear. That logic reproduced ChatGPT's Shopping behavior with about 95–97% accuracy in our classifier.
For brands selling physical consumer products, this is good news: Shopping visibility is predictable and optimizable. For brands in services, software, or financial products, Shopping is effectively a non-factor — but Answer Engine citation visibility is still very much in play.
Implications for AEO strategy
- Physical product brands should treat ChatGPT Shopping as a distinct optimization surface alongside organic citation visibility
- Product category is the primary lever — ensure your category taxonomy aligns with how ChatGPT classifies goods
- Constraint-rich product descriptions (price, features, compatibility) improve Shopping appearance rates
- Service brands should focus energy on citation optimization rather than Shopping mechanics