ChatGPT, Perplexity, and Google all launched AI shopping features in November 2025. Your customers are already using them. Here's what changes - and what to do about it.
ChatGPT launched Shopping Research on November 24, 2025. Perplexity and Google followed within days. The era of AI-powered shopping has arrived.
On November 24, 2025, ChatGPT launched Shopping Research - a feature that lets users ask "What's the best wireless earbuds under $200?" and get personalized recommendations with links to buy. OpenAI estimates ChatGPT is already handling roughly 50 million shopping-related queries per day. Perplexity launched a competing experience the next day. Google had rolled out expanded AI shopping features on November 13. The era of AI-powered shopping isn't coming. It arrived.
We're building a tool to check if your website is AI shopping ready. Jump to early access below.
Who this is for: Small business owners, ecommerce founders, and product-based entrepreneurs who want to understand how AI systems like ChatGPT and Perplexity recommend products - and how to optimize their websites to appear in these AI-driven shopping results.
For small business owners, this shift matters more than any algorithm change since Google launched. When customers ask AI for product recommendations, your business is either in the answer or invisible. There's no page two to hope for. No scrolling past ads. Just a direct recommendation - or silence.
I've spent the last week testing all three platforms, analyzing what makes businesses appear in AI recommendations, and documenting what actually works. This isn't speculation. It's field research.
The AI shopping race happened faster than most business owners realized. Here's the timeline:
Agentic checkout and AI Mode shopping rolled out ahead of the holidays. Product carousels, review synthesis, and direct purchase links integrated into Search.
Personalized buyer's guides, product comparisons, and curated recommendations with no sponsored placements in results today.
Buy directly within AI answers via PayPal integration. No ads, just curated recommendations based on user queries.
The timing wasn't coincidental. Black Friday 2025 became the first major shopping event where AI assistants were actively recommending products to millions of users. Some businesses saw this traffic. Most didn't know it existed.
Traditional search works on keywords. You optimize for "wireless earbuds under $200" and hope to rank. AI shopping works differently - it synthesizes information from multiple sources and makes recommendations based on context, reviews, specifications, and user preferences.
Here's what I found testing the same query across all three platforms:
| Platform | Sources Used | How It Recommends |
|---|---|---|
| ChatGPT | Product pages, expert reviews, Reddit discussions, manufacturer specs | Synthesizes pros/cons, creates personalized buyer's guide, asks follow-up questions |
| Perplexity | Live web data, review aggregators, retail sites, forums | Shows sources inline, offers "Buy" buttons, updates with real-time pricing |
| Google AI | Shopping graph, merchant data, reviews, video content | Product carousels with AI summaries, price comparisons, nearby availability |
The critical difference from traditional SEO: these systems read and understand your content, not just index it. A product page stuffed with keywords but lacking genuine information won't surface. A detailed product description with authentic reviews and clear specifications will.
After testing dozens of product queries across all three platforms, patterns emerged. In my testing, four things consistently mattered most for AI shopping visibility:
AI systems can't recommend what they don't understand. Vague descriptions like "high-quality materials" get ignored. Specific details like "14k solid gold, 2mm band width, comfort fit interior" get synthesized into recommendations.
This matters especially for small businesses with unique products. The AI can't compare your handmade ceramic mug to Amazon's if you don't explain what makes it different - the clay source, the glaze technique, the firing process.
All three platforms heavily weight customer reviews, but they read them - not just count them. A product with 50 detailed reviews explaining specific use cases outperforms a product with 500 generic "great product" reviews.
What I noticed: Reviews that mention specific problems solved ("finally found earbuds that don't fall out during running") appeared in AI responses almost verbatim.
AI shopping responds to questions like "What's the best gift for a wine lover?" or "What jewelry is good for sensitive skin?" Businesses that explain who their products are for - not just what they are - get matched to these queries.
Technical, but important. Product schema markup helps AI systems understand your inventory - prices, availability, specifications, reviews. Without it, you're harder to parse and less likely to appear.
Customers won't browse pages of results. They'll ask AI for recommendations and click the suggestions. Being in that first answer is the new "page one."
AI reads your content to understand your products. Thin descriptions, stock photos, and missing details make you invisible to these systems.
Not just star ratings - the actual text of reviews gets synthesized into recommendations. Encouraging detailed reviews pays off.
AI can match specific queries to specific products. "Sapphire jewelry for September birthdays" favors specialists over generalists.
Here's what's interesting: most small business owners don't know this shift happened. While enterprise brands scramble to optimize for AI shopping, independent businesses have a window to get ahead.
The advantage? AI systems are designed to favor useful information over brand size. A specialty jewelry business with detailed product descriptions, authentic reviews, and clear use cases can outperform a major retailer with generic listings.
The Gap: 50 million shopping queries daily on ChatGPT alone. Most small business websites aren't optimized to appear in these results. The businesses that adapt now capture this traffic before competitors realize it exists.
This isn't about overhauling everything. It's about making strategic improvements that help AI systems understand and recommend your products.
The good news: these improvements also help your AI content systems work better. When your product information is clear and detailed, AI can write better descriptions, answer customer questions more accurately, and represent your brand more effectively across every channel.
We're building a tool that analyzes your product pages and shows exactly what AI shopping systems see - and what's missing. Based on what's actually working for my 7-figure jewelry business, not theory.
AI shopping is in its first weeks. The platforms will evolve. New players will enter. But the fundamental shift - from keyword-based search to AI-synthesized recommendations - isn't going away.
The businesses that understand this early have an advantage. Not because they gamed an algorithm, but because they made their products genuinely understandable and useful to the systems customers are already using.
Your customers are asking AI what to buy. The question is whether AI knows to recommend you.
Google expanded AI Mode shopping on November 13, 2025. ChatGPT launched Shopping Research on November 24. Perplexity launched Instant Buy with PayPal on November 25. All three platforms launched within a 12-day window, right before Black Friday 2025.
Traditional search ranks pages based on keywords and links. AI shopping synthesizes information from multiple sources - product pages, reviews, forums, manufacturer specs - and provides direct recommendations. There's no "page two" of results. You're either in the AI's recommendation or invisible.
Yes. AI systems favor useful, specific information over brand recognition. A specialty store with detailed product descriptions, authentic reviews explaining specific use cases, and clear information about who products are for can outperform generic listings from major retailers. The advantage goes to businesses that genuinely help customers make decisions.
Four priorities: 1) Add detailed, specific product descriptions with exact specifications, 2) Encourage customers to leave detailed reviews explaining what problems your product solved, 3) Add context about who your products are for and when to use them, 4) Implement proper product schema markup so AI systems can parse your data.
69% of Google searches now end without a click. AI Overviews appear in 13%+ of queries. Here's what you need to understand.
Most entrepreneurs spend 8-12 hours weekly editing generic AI outputs. The difference is brand voice sophistication.
A clear comparison of n8n vs Zapier - who each is for, the pricing difference that matters.