Research

We scanned 21 well-known Shopify stores for AI-shopping readiness. The average grade was a C.

Published July 3, 2026 · SchemaRank Research

ChatGPT, Perplexity, and Google's AI Mode all read a product page's structured data (JSON-LD) to understand what it's selling — not the pretty HTML a human sees. We took 21 real, live, well-known Shopify-powered DTC brands and ran their product pages through the same 9-rule AI-readiness audit our free scanner uses. Here's what we found.

69/100
Average AI-Readiness score (Grade C)
10%
Had zero Product structured data at all
67%
Missing a GTIN or MPN identifier
62%
Missing a real aggregate rating in JSON-LD

Grade distribution

We scored each store on the same 0–100 scale as our public scanner: Offer/variant (15pts), description length (15pts), price, availability, brand, GTIN/MPN, aggregate rating, image, and description extractability (10pts each).

A
6/21 (29%)
B
7/21 (33%)
C
1/21 (5%)
D
4/21 (19%)
F
3/21 (14%)

Two stores — both large, well-funded DTC brands — had zero Product-typed structured data on their product pages at all. Not "incomplete." Zero. Their pages had other JSON-LD (breadcrumbs, organization info) but nothing describing the actual product being sold. To an AI shopping agent parsing structured data instead of rendering the page like a human, that product may as well not exist.

What's missing, most often

Product images and basic offer data are essentially solved — Shopify's own theme defaults handle those for almost everyone. The fields that are actually missing are the ones that make a product identifiable and trustworthy to an AI agent, not just visually presentable to a human:

GTIN or MPN (product identifier) 67% missing
Real aggregate rating in JSON-LD 62% missing
Description ≥ 50 words 43% missing
Positive price listed 19% missing
Product offer / variant 14% missing
In-stock status 14% missing
Brand or vendor 14% missing
Product image URL 10% missing

Rothy's, Gymshark, Nomad Goods, and Bulletproof Coffee all scored a perfect 100/100 in our sample — proof that full AI-readiness is achievable at scale on Shopify, not a theoretical ideal. It's not a platform limitation. It's a completeness problem, and it's fixable per-product in minutes, not a rebuild.

Methodology

We sampled 21 real, live, publicly-verifiable Shopify-powered DTC brands across apparel, beauty, food/beverage, and home goods, chosen for public name recognition rather than any prior relationship with SchemaRank. For each, we loaded one real product page and extracted every <script type="application/ld+json"> tag present at page load, then scored it against the same 9-rule engine used by SchemaRank's free public scanner. This checks only structured data present in the initial page load — some stores' review apps inject rating data client-side after load, which this method would not capture, so real review-data completeness may be modestly higher than reported here for a few stores. We did not access any store's admin, backend, or private data — everything scored here is publicly visible in any browser's "View Page Source." Two brands from our original candidate list could not be reliably reached during this research pass and were excluded rather than guessed at.

Curious about your own store?

Run the same 9-rule audit on your Shopify product page, free, in seconds.