By Steve Merrill, Founder of WRKNG Digital | November 4, 2026
Nine product feed fields decide whether an AI shopping assistant recommends your Shopify store: title, description, GTIN/MPN, brand, Google product category, price and availability, attributes and custom labels, images, and reviews. Get these clean and complete and AI can read you. Leave them thin and you are invisible, no matter how good the product is.
This is the field-by-field companion to our earlier guide. If you have not already, go test whether AI can actually read your feed first. This checklist tells you exactly what to fix once you find the gaps.
1. Product Title
The title is the first thing an AI assistant parses, and it carries the most weight for matching a shopper's question to your product. Google Merchant Center's title spec recommends front-loading brand, product type, and key attributes like size or color. Write for a human reading a one-line answer, not for keyword stuffing. "Stuffing" gets your items disapproved and ignored.
2. Product Description
AI reads the description to understand use cases, materials, and fit, which is how it decides if your product answers the actual question. Vague marketing copy gives it nothing to match against. Put the real specifics in the first 160 characters, because that is the slice most feeds and crawlers prioritize.
3. GTIN and MPN
These are the unique identifiers that let AI confirm your product is a real, verifiable item and not a mystery listing. Google's unique product identifier rules treat a correct GTIN as a trust signal, and missing GTINs are one of the most common reasons feeds get throttled. If a product has a barcode, the GTIN exists. Find it and add it.
4. Brand
Brand ties your product to a known entity, which AI uses to cross-reference reviews, authority, and reputation across the web. An empty brand field makes your item look generic and lowers how often it surfaces. Fill it on every single SKU, including your own private-label products.
5. Google Product Category
This field tells AI what shelf your product belongs on, and the wrong category means you show up for the wrong questions. Shopify's own product category taxonomy maps to Google's, so set it at the product level instead of letting an auto-guess run. Specific beats broad. "Women's Running Shoes" pulls better than "Apparel."
6. Price and Availability
AI will not recommend a product it thinks is out of stock or wrongly priced, and stale feed data is the fastest way to get dropped. Make sure your price and availability sync in real time, not on a daily batch. A $0 or mismatched price between your feed and your landing page is an instant trust break.
7. Attributes and Custom Labels
Structured attributes like color, size, material, age group, and gender let AI filter your product into precise answers like "waterproof hiking boots for wide feet." The more attributes you supply, the more specific questions you can win. Custom labels also help you group products for your own reporting without cluttering the public-facing fields.
8. Images
AI shopping tools increasingly read images, not just text, and a clean main image on a plain background is what gets parsed correctly. Google's image requirements call for a high-resolution shot with no promotional text or watermarks overlaid. Bad images get items disapproved before AI ever sees them.
9. Reviews and Rating
Reviews are social proof that AI weighs heavily when deciding which of ten similar products to actually recommend. Products with structured rating data get surfaced more often than identical products without it. Connect a review app that passes star ratings into your feed, and keep collecting them. This is the field most stores skip.
How We Chose This List
These nine fields come from the product data specs that AI shopping assistants pull from, including Google's Merchant Center requirements and Shopify's native feed structure. We ranked them by how often a missing or wrong value causes an item to be ignored or disapproved across the stores we audit.
FAQ
Which product feed field matters most for AI recommendations?
The title carries the most weight because it is the first field AI parses to match a shopper's question. But GTIN and availability are the fastest to break your listing if they are wrong.
Do I need a GTIN for every product?
Yes, if the product has a manufacturer barcode. For true one-of-a-kind or custom items with no GTIN, use the MPN and brand together so AI can still identify it.
How often should my Shopify feed update?
Price and availability should sync in real time or near it. Daily batch updates are too slow, and stale data gets your products dropped from AI answers.
Will good feed fields help if AI can't read my feed at all?
No. Clean fields only matter if the feed is accessible in the first place. Run the feed readability test before you optimize anything.
Do images really affect AI product recommendations?
Yes. AI shopping tools now read images, and a clean main image with no overlaid text is what gets parsed. A bad image can get the whole item disapproved.
Want us to audit all nine fields across your catalog and fix the gaps so AI starts recommending you? See how we do it at WRKNG Digital's agentic commerce service.
