Product specs and comparisons AI shoppers can read: materials, dimensions and 'X vs Y' for the holidays

Updated
Reading time
6 min
Sources
10 linked
For AI tools
Markdown version

Short answer

AI shoppers compare products by reading facts: what it's made of, how big it is, what it works with, and how it differs from the item next to it. If those facts sit in a photo, a PDF or a vague paragraph, an assistant may skip your product or get it wrong. Write specs as short "name: value" lines on the page, put the same facts into your store's attribute fields and your Google and ChatGPT feeds, and answer "which one should I buy?" yourself in plain text.

For brand, titles and descriptions, start with our holiday product data checklist. This guide covers specs and comparisons.

Why comparisons go wrong

Shoppers ask things like "which of these two blankets is warmer?" or "will this case fit my phone?" An assistant can only answer from what you've written. Common gaps:

  • Dimensions only in a product photo
  • Material described as "premium" instead of named
  • Compatibility only in reviews or a support page
  • Two similar products with the same description, so nothing tells them apart (see the problems we see most often)

Write specs as name: value pairs

The simplest format works everywhere: one fact per line, with a label and a value.

  • Material: 100% wool
  • Dimensions: 30 x 20 x 12 cm
  • Weight: 0.75 kg
  • Works with: [list the exact models]
  • Care: hand wash cold

Use the same units and wording on the page, in your attribute fields and in your feeds. Sizes and colors have their own rules, covered in naming sizes and colors and size charts as text.

Shopify: category metafields

Shopify's category metafields are product attributes tied to a category in Shopify's Standard Product Taxonomy. Shopify says they help make products "more discoverable by visitors on your site, on marketplaces, and on search engines." For a shirt, for example, you can add size, neckline and sleeve length type.

Start by picking the right product category for each product. Then review the category metafields it suggests and fill in the ones that matter for comparison. Shopify notes that some are added by default based on AI predictions, so check them before you trust them.

WooCommerce: product attributes

WooCommerce product attributes can be global (reused across many products) or custom to one product. WooCommerce says they help organize and filter products and are needed for variable products. Use global attributes for specs you sell across many items, such as material, so the wording stays the same everywhere. For how filters behave with AI shoppers, see filters: useful, with limits.

Google Merchant Center: details, highlights and material

  • Product detail (product_detail) holds technical specs as a section name, attribute name and value, such as "Warranty:1 year manufacturer". Google says to include only confirmed values, not keywords, and not price, shipping or your company name.
  • Product highlight (product_highlight) is a short list of the product's most important points. Google requires at least 2 if you use it, recommends 4 to 6, and says to describe only the product itself, not compatible products or accessories. So keep "works with" lists in product details and on the page.
  • Material (material) is required for products that vary by material and recommended when shoppers may choose by material. Put the main material only, not color or size.
  • GTIN and MPN identify the exact product. Google says products with a missing or incorrect GTIN may have limited visibility, and also says don't guess or make up a GTIN or MPN.

ChatGPT: material, dimensions and weight

OpenAI's product feed spec has optional fields for material ("principal materials in the item"), a structured dimensions field with a unit, and weight with item_weight_unit. OpenAI says to use dimensions for new feeds, and to leave dimensions out if you don't know them. We found no general spec or compatibility field, so those details need to be in your title, description or page text.

Write the "X vs Y" answer yourself

The surest way to get a comparison right is to publish it. On each product that has a close sibling, add a short plain-text section:

  • Name both products exactly as they're titled.
  • List the real differences as name: value lines, such as material, size, weight and what it works with.
  • Say who each one suits, using facts only ("lighter, for travel" rather than "best ever").

Use a simple HTML table or list, not an image. If you sell bundles, list the contents the same way, as covered in say exactly what's inside.

What the store's own tools see

On Shopify, the built-in WebMCP tools let an agent in a supported browser call search_catalog to search products, get_product for "full details for a product, including variants, prices, and which option combinations are in stock", and show_variant to open the page with a specific option selected. Clear specs on the product make those answers better. See the default tools.

Holiday spec and comparison checklist

  • Key specs written as name: value lines in page text, not only in images or PDFs
  • Same units and wording on the page, in attributes and in feeds
  • Shopify: right product category set, category metafields checked and filled in
  • WooCommerce: global attributes used for shared specs like material
  • Merchant Center: product_detail for specs, 4 to 6 product_highlight lines, material where it matters
  • GTIN and MPN only where you're sure they're correct
  • ChatGPT feed: material, dimensions and weight filled in where known
  • An "X vs Y" section on products with a close sibling

FAQ

What's the difference between product details and product highlights in Merchant Center? Details are structured specs (name and value). Highlights are short sentences about the product's most important points.

Can I list compatible models in product highlights? Google says highlights should describe only the product itself, not compatible products. Put compatibility in your product details and page text.

Does ChatGPT's feed accept dimensions? Yes. OpenAI's spec has an optional dimensions field with length, width, height and a unit, plus weight.

Should I make up a GTIN if my product doesn't have one? No. Google says not to guess or make one up.

Do I need a comparison table for every product? No. Start with products shoppers often mix up, such as two sizes or two models of the same item.

Get a free product data check

Not sure your specs read correctly to AI shoppers? We'll check your specs, attributes and feed fields, and list what to fix. See a sample report first, or run a free scan. Questions? Email hello@convertwebmcp.com.

Request a free product data check

Sources

  1. Google Merchant Center Help: Product detail [product_detail]
  2. Google Merchant Center Help: Product highlight [product_highlight]
  3. Google Merchant Center Help: Material [material]
  4. Google Merchant Center Help: GTIN [gtin]
  5. Google Merchant Center Help: MPN [mpn]
  6. Shopify Help Center: Category metafields
  7. Shopify Help Center: Product category (Standard Product Taxonomy)
  8. Shopify Dev: WebMCP
  9. WooCommerce Docs: Managing Product Categories, Tags and Attributes
  10. OpenAI: Product feed spec
Next guideHoliday free shipping thresholds and shipping costs AI shoppers can quote correctly

Free product data check for ChatGPT, AI shoppers and BFCMSee a sample report (EXAMPLE)

Get a free scan.

Black Friday starts November 27. Free scan — what AI shopping assistants can do on your store, what's missing, and three tools with suggested code. We email your free report.

If you want us to implement the fixes after your report, ask — builds stay paid only when you ask.

Free report by email. No login. No admin access.

See a sample report (EXAMPLE)

YOUR STORE · PRODUCT DATA CHECK · FREE