What product fields does AI shopping need?

AI shopping relies on clear titles, accurate prices, real-time availability, unique identifiers like GTINs, quality images, and concise descriptions. Missing or inconsistent fields are the most common reason products are excluded — the checker catches them.

How is this different from a Google Merchant feed check?

Merchant feed checks target Google Shopping rules. This checker validates against what AI shopping assistants need to recommend a product confidently — overlapping but distinct, with emphasis on completeness and clarity for AI.

Why are my products not showing in AI shopping?

Common causes include missing identifiers, stale availability, vague titles, or feed errors that prevent the AI from trusting the listing. The checker surfaces each issue so you can fix the right one.

Do I need a full feed file to use it?

No. You can validate any feed URL directly. Enter the URL and get your full field-coverage analysis in seconds.

How does AI decide which products to recommend in shopping answers?

AI shopping engines match buyer intent (the query) to product records they can trust. They look for a complete, consistent product record: a descriptive title that matches the query, an accurate price, real-time availability, a unique identifier like a GTIN, a quality image, and a factual description. Products with strong structured data and no conflicting signals between the feed and on-page schema score highest for recommendation confidence.

What is the difference between a product feed and on-page schema markup?

A product feed is a separate file (XML, CSV, JSON) submitted to channels like Google Merchant Center. On-page schema markup is JSON-LD embedded in your product page HTML that AI crawlers read directly. Both need to be consistent — many retailers maintain one carefully but let the other drift, creating the split reality that causes AI shopping exclusion.

Can I validate a single product page instead of my whole feed?

Yes. The ChatGPT Shopping Feed Checker will support single product page validation as well as feed URL input. This lets you spot-check high-priority products — bestsellers, new launches, seasonal items — before investing in a full catalog audit.

Do I need a GTIN for every product to appear in AI shopping?

GTINs are strongly recommended for any branded product that has one assigned. For custom or handmade products without a barcode, MPN and brand together can substitute as an identifier. However, AI shopping confidence is materially higher for products with GTINs because the engine can resolve the item unambiguously across multiple sources.

Why does my product appear in Google Shopping but not in ChatGPT shopping?

Google Shopping validates feed format and policy compliance. AI shopping engines apply a different layer of trust signals: factual descriptions (no promotional language), strong identifiers, consistent title and description across feed and product page, and real-time availability. A feed that passes Merchant Center validation can still fail AI shopping eligibility on these criteria.

How often should I validate my product feed for AI shopping?

Validate after every feed update, product launch, or price change. For large dynamic catalogs, a weekly automated validation is the minimum. Availability status in particular changes constantly and should be refreshed in near-real time — stale availability is the fastest path to AI shopping de-prioritization.

Does product description length matter for AI shopping?

Yes, but more is not better. AI shopping engines prefer descriptions of one to three sentences that are purely factual: what the product is, what it does, what it is made of, and who it is for. Long promotional or keyword-stuffed descriptions are treated as a negative trust signal. Aim for clarity and specificity over length.

Will fixing my feed for AI shopping also improve my Google Shopping performance?

In most cases, yes. The improvements required for AI shopping eligibility — complete identifiers, accurate availability, descriptive titles, factual descriptions — also strengthen Google Shopping performance by reducing data quality issues, improving match confidence, and reducing the chance of disapprovals. The two requirements overlap significantly, with AI shopping adding a layer on top.

Free Tool

Free ChatGPT Shopping Feed Checker

Validate your product feed for ChatGPT shopping results and make sure your products are eligible to appear when shoppers ask AI what to buy.

Supports Google Merchant Center XML and TSV/CSV formats.

What is a ChatGPT shopping feed?

AI shopping experiences pull from structured product data — titles, prices, availability, images, identifiers, and descriptions. If that feed is incomplete or malformed, your products can be skipped when ChatGPT and other AI assistants recommend what to buy.

The ChatGPT Shopping Feed Checker validates your product feed against the fields and formats AI shopping needs, flags missing or invalid attributes, and shows what to fix so your catalog is eligible to surface in AI-driven shopping answers.

How it works

Add your feed

Paste your product feed URL or product page and we read its structured data.

We validate the fields

We check titles, prices, availability, GTINs, images, and required attributes for AI shopping.

Get your fix list

See exactly which products and fields need work so your catalog is AI-shopping ready.

Frequently asked questions

The Complete Guide to AI Shopping Feeds: Getting Your Products Into ChatGPT Answers

AI shopping experiences work by pulling structured product data directly from feeds and web product pages — not by browsing your store like a human would. When a shopper asks ChatGPT or Perplexity "where can I buy a waterproof running jacket under $100", the engine retrieves product records it can parse: titles, prices, availability, images, GTINs, and descriptions formatted in schema.org Product markup. If your product data is machine-readable, complete, and consistent, your items become candidates for AI shopping answers. If it is partial, malformed, or missing key identifiers, the engine cannot surface your products with confidence and skips them in favor of catalog entries that are fully defined.

The product attributes AI shopping relies on go beyond what many retailers bother to populate. A clear, keyword-rich title that includes brand, model, material, and size is not optional — it is the primary field the engine uses to match buyer intent. Price must be accurate and current; stale or misleading prices cause the AI to distrust the whole listing. Availability must reflect real-time stock status — "in stock", "out of stock", or "pre-order" — because recommending an unavailable product damages the assistant's credibility. GTINs (Global Trade Item Numbers) such as EAN or UPC codes and MPNs (Manufacturer Part Numbers) are the fingerprints that let AI engines de-duplicate listings, compare prices across sellers, and confirm product identity without ambiguity. High-resolution images on a clean background, a factual description free of promotional language, and explicit brand and manufacturer fields round out the set of attributes that turn a product listing into a trustworthy AI shopping source.

Incomplete or inconsistent product data is the single most common reason catalogs are excluded from AI shopping answers. Retailers who update Google Shopping feeds occasionally but never audit schema.org markup on their product pages create a split reality: the feed validator passes, but the on-page structured data that AI crawlers actually read is stale, missing price information, or lacks a GTIN entirely. Consistency matters just as much as completeness — a product whose title says "Blue Merino Crew-Neck Sweater, Medium" in the feed but "Men's Sweater" on the product page gives the engine conflicting signals and lowers confidence. Even small gaps compound across thousands of SKUs, quietly eroding AI shopping eligibility at scale.

AI shopping requirements differ from traditional Google Shopping feed requirements in important ways. Google Merchant Center is primarily concerned with feed format compliance, prohibited content policies, and price accuracy for ad bidding. AI shopping assistants apply a richer set of trust signals: they care whether a product is clearly identifiable as a specific real-world item (hence the emphasis on GTINs and MPNs), whether the description is factual and useful rather than promotional, whether the brand is a recognized entity, and whether the product page reinforces what the feed says. The ChatGPT Shopping Feed Checker is designed specifically for this AI-native layer — it validates the attributes, consistency, and trust signals that determine whether AI assistants include or exclude your products from their shopping recommendations.

AI Shopping Feed Optimization Playbook

Eight actionable steps to make your product catalog fully eligible for AI shopping recommendations.

Audit feed completeness across all required fields

Pull your entire feed and calculate fill rates for every required field: title, price, availability, GTIN/MPN, main image, description, brand, and condition. Aim for 100% on identifiers and availability; missing values in even a small share of SKUs can disqualify your whole catalog from AI shopping placement.

Ensure every product has a unique identifier

GTINs (EAN, UPC, ISBN, JAN) and MPNs are the primary signals AI uses to resolve a product as a specific real-world item. Without them, the engine cannot de-duplicate your listing against other sellers, cannot confirm price accuracy, and treats the product as unverifiable. Source GTINs from GS1 or your manufacturer; add MPN for every variant.

Write descriptive titles that include key product attributes

Your title should follow the pattern: Brand + Product Type + Key Attributes (color, material, size, model). "Nike Men's Air Zoom Pegasus 40 Running Shoe, Black, Size 10" gives the AI everything it needs to match buyer intent precisely. Avoid vague titles like "Men's Shoe" — they reduce match confidence and lower AI shopping eligibility.

Set real-time availability status on every product

Use schema.org availability values — InStock, OutOfStock, PreOrder, BackOrder — and keep them synchronized with actual inventory in near-real time. AI assistants that recommend an out-of-stock product to a buyer lose credibility; they protect against this by deprioritizing feeds where availability accuracy is uncertain.

Add quality product images that meet AI shopping standards

Use high-resolution images (at minimum 800x800px) on a clean, uncluttered background. Show the product clearly, without watermarks or promotional text overlaid. AI shopping surfaces images alongside product recommendations; poor-quality or heavily branded images reduce the trustworthiness of the listing and can prevent inclusion.

Write clear, factual descriptions — no promotional language

AI engines treat promotional copy ("Best in class!", "Limited time offer!") as a trust signal against the listing. Write descriptions that specify what the product is, what it does, what it is made of, and who it is for. One to three focused sentences that add information the title does not already contain. Factual, concise, and free of superlatives.

Ensure brand and manufacturer fields are populated and consistent

The brand field should match your canonical brand name exactly — the same string used in your Organization schema, on G2, Trustpilot, and your Google Business Profile. Manufacturer should reflect the actual maker of the product. Inconsistencies between feed, product page, and external records create entity confusion that reduces AI shopping eligibility.

Validate with the ChatGPT Shopping Feed Checker before going live

After every feed update or product launch, run your feed URL or key product pages through the ChatGPT Shopping Feed Checker. It surfaces field-level errors, consistency gaps between feed and on-page schema, missing identifiers, and availability mismatches — giving you a prioritized fix list before the next AI crawl picks up your data.

AI Shopping Feed Glossary

Key terms for product feed optimization and AI shopping eligibility, in plain language.

Product Feed: A structured file (XML, CSV, or JSON) or on-page markup that contains your product catalog data — titles, prices, availability, images, identifiers, and descriptions — in a format machines can read and ingest.
GTIN: Global Trade Item Number — a standardized numeric identifier assigned to a physical product, including EAN-13 (Europe), UPC-A (North America), ISBN (books), and JAN (Japan). GTINs allow AI shopping engines to uniquely identify a product and compare listings across sellers.
MPN: Manufacturer Part Number — a unique code assigned by the manufacturer to identify a specific product model or variant. MPNs complement GTINs for products without a retail barcode, particularly in electronics and industrial goods.
Schema Product: The schema.org/Product structured data type, added to product pages as JSON-LD or microdata. It exposes product attributes (name, offers, brand, identifier, image, description) directly to crawlers in a machine-readable format that AI shopping engines ingest alongside feed data.
Availability Status: A schema.org or feed value declaring whether a product is currently purchasable: InStock, OutOfStock, PreOrder, BackOrder, or Discontinued. Accurate real-time availability is critical for AI shopping inclusion because assistants will not recommend products they cannot verify are purchasable.
Structured Product Data: Any machine-readable product information — whether in a feed file or on-page schema markup — that allows search and AI engines to understand a product's attributes without having to parse natural language.
Feed Validation: The process of checking a product feed or on-page schema against a defined set of rules — required fields, valid formats, logical consistency — to identify errors before they reach search and AI shopping engines.
AI Shopping: The experience where an AI assistant recommends specific products with titles, prices, images, and purchase links in response to a natural language query — a format increasingly offered by ChatGPT, Perplexity, Google, and Bing.
Shopping Action: A structured recommendation from an AI assistant that includes a specific product, its price, a merchant, and a direct link to purchase — the AI equivalent of a Google Shopping ad unit.
Product Identifier: Any code or number that uniquely identifies a product — GTIN, MPN, SKU, or model number. Strong product identifiers allow AI engines to resolve exactly which real-world item you are selling and match your listing to buyer intent with high confidence.
Rich Product Result: An enhanced search or AI result that displays product attributes — image, price, rating, availability — alongside the title and link. Rich results require complete, valid schema.org Product markup and are the visual form AI shopping recommendations take.
Dynamic Pricing: A retail strategy where product prices change based on demand, competition, or time — which creates feed freshness challenges. AI shopping engines deprioritize feeds where prices are frequently stale relative to the actual checkout price, so real-time feed updates are essential for dynamic pricers.

Who Needs a ChatGPT Shopping Feed Checker

Any retailer or agency whose products should appear when buyers ask AI what to buy.

E-commerce brands wanting AI shopping placement

Direct-to-consumer and mid-market retailers whose products are eligible for AI shopping recommendations but are not appearing. The checker identifies the specific field gaps and consistency issues preventing inclusion — missing GTINs, stale availability, vague titles — and prioritizes what to fix first.

Fashion and apparel retailers with complex variant data

Apparel catalogs with hundreds of color and size variants are especially prone to incomplete identifier data and inconsistent titles across variants. The checker flags variants missing GTINs, availability not set per variant, and title patterns that do not include the attributes buyers search for.

Electronics retailers with strict GTIN requirements

Consumer electronics are among the categories where GTINs and MPNs are most strictly required by AI shopping engines. Missing or incorrect identifiers cause electronics listings to be skipped entirely. The checker verifies identifier completeness and consistency with on-page schema markup.

D2C brands building AI-native product discovery

Brands launching or growing with AI shopping as a primary channel need their product data structured for AI from day one. The checker provides a baseline audit before launch and a recurring validation check so that new product additions never ship with feed errors.

Multi-channel retailers syncing feeds across platforms

Retailers selling on their own site, Amazon, Google Shopping, and comparison engines often have feed data that diverges across channels. The checker validates the on-site feed and schema markup independently, catching inconsistencies that platform-specific validators miss.

Agencies auditing client product feeds

E-commerce SEO and performance agencies that manage product data for multiple clients can use the checker to produce a fast, structured feed audit for any URL. The output is a prioritized list of issues by severity — identifiers, availability, titles, descriptions — that maps directly to a client deliverable.