By Steve Merrill, Founder of WRKNG Digital — June 21, 2026
AI shopping assistants read your product feed before they recommend anything. If your feed is missing these seven fields, you're invisible — no matter how good your product actually is.
ChatGPT Shopping, Google AI Overviews, and Perplexity all pull structured product data to match queries to products. They don't browse your store. They parse your feed. These are the fields that determine whether you show up.
1. GTIN / Product Identifier
A GTIN (Global Trade Item Number) is how AI shopping assistants cross-reference your product against external databases, price comparison engines, and review aggregators. Without one, ChatGPT Shopping and Google AI Overviews can't confirm your product is what you say it is — so they often won't recommend it at all.
Google's Merchant Center product data specification lists GTIN as required for all branded products. This isn't optional. It's the handshake between your catalog and the AI's knowledge graph.
2. Detailed Product Type
The product_type field is where most Shopify stores take a shortcut that costs them. Broad entries like "Clothing" or "Home Goods" don't give AI assistants enough signal to match your product to a specific query. "Women's Merino Wool Running Shorts" does.
According to Salsify's 2024 Product Content Benchmark Report, products with seven or more feed attributes convert 40% better than those with fewer than three. Specificity is the whole game.
3. Long Description (300+ Words, Attribute-Rich)
Most Shopify product descriptions are written for humans skimming a page. AI assistants aren't skimming — they're parsing for attributes: materials, dimensions, use cases, care instructions, compatibility. A description like "comfortable running shorts" tells an AI almost nothing.
"Lightweight Merino wool shorts, 4-inch inseam, UPF 50+ protection, reflective trim, suitable for trail and road running in 40–70°F conditions" — that's something an AI can match to a query. I've audited over 60 Shopify stores this year. The description field is consistently the most underdeveloped asset in every single feed.
4. Material / Composition Attributes
The [material] or [fabric] field is used directly by AI shopping assistants when users run queries like "wool sweaters under $100" or "BPA-free kids' water bottles." If it's not in your feed, you can't match those queries — full stop. No amount of on-page text compensates for a missing feed attribute.
This applies well beyond apparel. Cookware, toys, furniture, beauty — every category has material-based queries that AI assistants now answer with product recommendations pulled directly from structured feed data.
5. Product Highlights (feature_description)
Google added feature_description as an optional field in 2023, but it carries significant weight in AI Overviews. It maps directly to how AI presents products — as bullet-point facts, not paragraphs. Each highlight should be one complete factual statement: "Waterproof to 100m, tested to ISO 22810:2010." Not "waterproof."
Aim for 4–6 highlights per product. One sentence each. Treat it like a spec sheet, not marketing copy. AI doesn't reward enthusiasm — it rewards precision.
6. Aggregate Rating (Review Score + Count)
AI shopping assistants don't just look at your star rating. They look at volume too. A 4.9 with 12 reviews is weaker than a 4.6 with 840 reviews — and the AI models know the difference. Review count is a trust proxy, not just a vanity metric.
Shopify apps like Judge.me and Okendo both support direct product feed integration for aggregate rating data. If your review data isn't structured and in your feed, assume it's not being factored into AI recommendations. Assume nothing carries over automatically.
7. Availability + Fulfillment Speed
"In stock" isn't enough anymore. AI shopping assistants — especially Google's — increasingly filter by fulfillment speed, particularly for time-sensitive queries. Google's ships_from_country, min_handling_time, and transit_time attributes feed directly into AI-powered Shopping filters and recommendations.
A 2024 Tinuiti Benchmark Report found that products with accurate availability and handling time data saw 23% higher click-through rates from Shopping surfaces compared to products with incomplete fulfillment data. Accurate beats optimistic every time.
How We Chose This List
These seven fields come from cross-referencing three sources: Google's official Merchant Center product data specification, Salsify's 2024 product content research, and direct feed audits of Shopify stores in the WRKNG Digital client base.
The filter was simple: which fields appear consistently in AI-generated product recommendations and Shopping surfaces? These seven showed up every time. Fields like color and size matter for ads but don't move the needle on AI recommendation surfaces the same way. These do.
Frequently Asked Questions
Do I need all seven fields, or just a few?
All seven. They're not a pick-three situation. Each field answers a different kind of query. Missing one means missing an entire class of searches — all the queries that depend on that attribute to return a match.
Where do I add these fields in Shopify?
Most of these fields live in your Google Merchant Center feed, not your Shopify product pages directly. You can manage them through Shopify's Google & YouTube channel app, a third-party feed tool like DataFeedWatch or Feedonomics, or by editing your feed source directly in Merchant Center. The description and material fields can be set in Shopify metafields and mapped into your feed from there.
How often should I update product feed data?
Availability and price should update at least daily — ideally in real time. Everything else (descriptions, materials, highlights, GTINs) should be reviewed quarterly or whenever product specs change. Stale data creates mismatches that AI assistants penalize by dropping your placement.
Does this apply to stores using Shopify Markets or international storefronts?
Yes, but each market needs its own feed. You can't submit one English-language feed with USD pricing and expect AI shopping assistants in the EU or UK to surface your products accurately. Language, currency, and availability need to match the target market in each feed submission.
Can AI shopping assistants see my Shopify store directly without a feed?
Some can crawl your site for schema markup, but structured product feeds remain the primary data source for ChatGPT Shopping and Google AI Overviews. Relying on crawl-only discovery is slower, less reliable, and gives you no control over how your data is presented. Feed-first is the right approach in 2026.
Want to See How Your Feed Scores?
Most Shopify stores are missing at least four of these seven fields. We built a tool that audits your product feed against the exact criteria AI shopping assistants use to evaluate and recommend products.
