OpenAI has introduced Shopping Research, a new feature in ChatGPT designed to help users identify suitable products without checking multiple websites. Users can describe what they need—for example, “Find the quietest cordless stick vacuum for a small apartment,” “Help me choose between these three bikes,” or “I need a gift for my four-year-old niece who enjoys art”—and the system produces a detailed, customized guide within minutes.
Purpose of Shopping Research
The company noted that millions of people use ChatGPT to understand and compare products. Shopping Research supports more detailed decision-making by helping users evaluate options based on preferences, features, and budget. It performs particularly well in categories such as electronics, beauty, home and garden, kitchen and appliances, and sports and outdoor.
For straightforward queries—like verifying a specification or checking a price—a regular ChatGPT response is sufficient. When comparisons, constraints, or tradeoffs are involved, Shopping Research provides structured, in-depth results.
How It Works
1. Starting the Research
Users can begin by asking a shopping-related question or selecting “shopping research” from the (+) menu. A visual interface appears, allowing users to explain their requirements and respond to clarifying questions about budget, intended user, and feature priorities. If ChatGPT memory is active, prior interactions can influence suggestions—for instance, gaming-related discussions can shape laptop recommendations.
2. Guiding the Research
In the background, the system gathers updated information across the internet, including prices, availability, specifications, reviews, and images. Users can guide the process by selecting “Not interested” or “More like this”, enabling real-time refinement.
3. Receiving a Buyer’s Guide
After completing its research, ChatGPT generates a personalized buyer’s guide outlining recommended products, key differences, tradeoffs, and verified retailer data. Users can visit retailer sites directly from the guide. A future update will support purchases through Instant Checkout for participating merchants.
4. Integration with ChatGPT Pulse
Shopping Research is also available in ChatGPT Pulse for Pro users. Pulse can surface relevant buyer’s guides automatically based on past discussions—for example, recommending e-bike accessories after an e-bike conversation.
Technology and Accuracy
Shopping Research is built on a version of GPT-5 mini, trained with reinforcement learning for shopping-specific tasks. The model synthesizes information from reliable online sources, cites them, and updates results dynamically. Product accuracy is evaluated based on how many suggested items meet user-defined requirements such as price, material, color, and specifications.
Transparency and Limitations
- User chats are not shared with retailers.
- Results come from publicly available retail sites, with citations and filtering of low-quality or spam sources.
- The system may occasionally present incorrect details related to price or availability; users are advised to confirm information on retailer sites.
Availability
Shopping Research is rolling out on mobile and web for ChatGPT users across Free, Go, Plus, and Pro plans. During the holiday period, usage is nearly unlimited for all supported plans.