Product Discovery Is Splitting Into Two Channels
For the first time since Google Shopping launched, product discovery has a second major channel: AI-powered shopping. Buyers are asking ChatGPT "what's the best espresso machine under $500," asking Perplexity to compare specific models, and getting product recommendations directly inside Google AI Overviews. These are not future features on a product roadmap — they are live and handling real purchase-intent queries today. The shift happened faster than most ecommerce operators expected, and it is accelerating.
What makes this different from every previous "new channel" in ecommerce is that AI shopping surfaces synthesize answers rather than listing links. A buyer who asks Perplexity to compare two standing desks gets a structured answer with product cards, prices, and buy links — not ten blue links to click through. The experience is fundamentally different from traditional search, and so is the path a store must take to be visible in it. Stores that are not visible in these answers are missing a growing share of the buyer journey, and that share is growing every quarter.
The good news: the strategies that earn visibility across AI shopping surfaces overlap significantly. Structured product data, authoritative content, and clean technical access work on all of them. The stores that invest now will compound their visibility across every AI surface simultaneously.
Perplexity Shopping
Perplexity has built the most developed AI shopping experience of any platform. When a buyer searches for a product category or specific product, Perplexity displays dedicated product cards with images, current prices, availability status, and direct buy links — all embedded alongside a cited text answer that explains the recommendation. This is not a search engine that links to shopping results; it is a shopping surface that integrates product data into conversational answers.
Perplexity pulls product data from two sources: merchant feeds (structured data submitted by retailers) and Product schema markup on individual product pages. Price, availability, brand, and image data in your schema directly feeds what Perplexity shows in product cards. Stores with complete, accurate Product schema are significantly more likely to appear in Perplexity's shopping results than stores with minimal or missing structured data.
Critically, Perplexity also cites content pages alongside product cards. A buying guide, a detailed review, or a comparison page on your site can earn a Perplexity citation even when the product cards come from a different retailer. This means both your product pages (via schema) and your content pages (via authority and specificity) can drive visibility. The dual investment — structured product data plus authoritative content — is not optional; it is how the platform works.
ChatGPT Product Search
ChatGPT Search handles product queries with a mix of synthesized recommendations and cited sources. When a user asks "best running shoes for wide feet," ChatGPT provides a curated answer — typically recommending 3 to 5 specific products — and cites the sources it drew those recommendations from. The cited sources are usually review sites, authoritative buying guides, and brand pages that contain specific, testable product information.
The key distinction with ChatGPT is that product pages alone are rarely cited. ChatGPT's retrieval system favors content pages that compare, recommend, and evaluate products over individual product detail pages. A store's product page for "Model X Running Shoe" is unlikely to be cited when a buyer asks for the best running shoes. But a store's buying guide titled "Best Running Shoes for Wide Feet: 7 Models We Tested" is exactly the kind of content that earns ChatGPT citations.
This means the content strategy for ChatGPT visibility is content-first, not product-first. Stores need authoritative editorial content that references their own products with specific claims, measurements, and comparisons. The content earns the citation, and the citation drives the buyer to the store, where the product page closes the sale. It is a two-step path: content earns discovery, product pages earn conversion.
Google AI Overviews and Shopping
Google AI Overviews appear above traditional search results for an expanding set of product queries, synthesizing answers from web sources into a single panel that users can read without clicking through to any website. For commercial queries, these Overviews increasingly include product cards with prices, star ratings, and direct links — pulled from Google Shopping data via Merchant Center feeds. The result is a shopping experience embedded directly in the search results page.
Being visible in Google AI Overviews requires a dual strategy. First, your products need to be in Google Merchant Center with accurate, complete feed data — this is how product cards with prices and ratings appear in Overviews. Second, your content pages need to be authoritative enough to be cited as sources in the Overview's text synthesis. A store that has strong Merchant Center data but no content will get product cards but no text citations. A store with great content but no Merchant Center feed will get text citations but no product cards.
The winning combination is both: structured product data in Merchant Center that powers product cards, plus authoritative content pages (buying guides, reviews, comparison pages) that earn text citations in the Overview. This dual presence — appearing in both the product card section and the cited sources section — maximizes visibility and establishes the store as both a place to buy and a trusted authority on the product category.
What AI Shopping Surfaces Want from Your Store
Across all four major AI shopping surfaces, the requirements converge on three categories. First: structured product data. Every product page needs complete Product schema with accurate price, availability, brand, images, and ratings. Your Google Merchant Center feed should be active and current. Product URLs should be clean, crawlable, and stable — not session-based or parameterized. This structured data is the raw material that AI surfaces use to build product cards and verify product claims.
Second: authoritative content. Buying guides that recommend specific products with reasons. Comparison pages that evaluate products against each other on specific dimensions. Reviews with test data, measurements, or experience-based assessments. This content is what AI surfaces cite when synthesizing answers. Without it, your store has products but no voice — AI surfaces will cite someone else's content about your product category, and that someone else captures the buyer's trust.
Third: technical access. AI crawlers (GPTBot, PerplexityBot, Google-Extended, Bingbot) must be allowed in robots.txt. Pages must load fast and render content in clean HTML — not hidden behind JavaScript frameworks that require client-side rendering, login walls, or interstitials. If AI crawlers cannot access and parse your content, none of the other investments matter. Check your robots.txt today: if you are blocking any AI crawler, you are opting out of the fastest-growing discovery channel in ecommerce.
The Content Strategy for AI Shopping Visibility
Build three content types specifically for AI shopping visibility. First: "Best [category] for [use case]" guides that recommend and compare products in your catalog. "Best standing desks for small apartments," "Best wireless earbuds for running," "Best organic dog food for senior dogs." These guides match the exact query patterns buyers use in ChatGPT and Perplexity. They earn citations because they answer the question directly, cite specific products with reasons, and come from a store with domain expertise. Build one for every major product category and use-case intersection in your catalog.
Second: "[Product A] vs [Product B]" comparison pages for your top-selling products against competitors. "Aeropress vs French Press," "Dyson V15 vs Shark Stratos," "Allbirds Tree Runners vs Nike Pegasus." These match the comparison queries that buyers increasingly ask AI surfaces — queries that trigger AI answers at very high rates. Structure them with clear dimensions (price, features, durability, best-for), a verdict, and links to the products in your store.
Third: buyer's guides with specific criteria, test data, and ranked recommendations. Not generic "10 things to consider" listicles — substantive guides that demonstrate expertise. Include specific measurements, compatibility details, longevity assessments, and use-case recommendations. These guides serve as the authoritative source AI surfaces cite when synthesizing answers. The more specific and evidence-based your guide, the more likely it is to be cited over a competitor's generic content. This is where your content engine earns its return.
What This Means for Ecommerce Strategy
AI shopping is not replacing Google Shopping or organic search. It is adding a third discovery channel alongside paid ads and organic search. The buyer journey that used to be "Google search, click a few links, buy" is becoming "ask AI for recommendations, verify on Google, buy." Some buyers skip the verification step entirely and buy directly from AI-recommended sources. The discovery moment is shifting upstream, and stores that are not visible in that upstream moment lose the buyer before traditional search even enters the picture.
The stores that win will be visible in all three channels: paid ads for immediate purchase intent and brand defense, organic search for information-seeking and category browsing, and AI shopping for research, comparison, and recommendation queries. This is not three separate strategies — the content investments for AI shopping visibility (structured data, authoritative guides, comparison content) also improve organic search performance. There is no trade-off between optimizing for AI and optimizing for Google. The strategy is additive.
The cost of inaction is measurable and growing. Every quarter that passes without AI shopping visibility is a quarter where competitors who invested in structured data and authoritative content are compounding their advantage. AI surfaces are building increasingly sophisticated shopping features — product cards, price comparisons, inventory checks, direct checkout. The stores that are visible today will have established authority and citation history by the time these features mature. The stores that wait will be playing catch-up against entrenched incumbents. The time to invest is now, and the investment pays dividends across every discovery channel simultaneously.