Reports library
Catalog audits of real Shopify stores
Each report reads a store's public catalog the way Google Merchant Center and AI shopping agents like ChatGPT do, scores it, and lists every issue with what it costs. Open one to see exactly what an audit of your store looks like.
- Stores audited
- 5
- Products scanned
- 3,998
- Median ad feed health
- 83
- Median AI readability
- 56
Full report with fixes
brooklinen.com
Brooklinen
Ad feed health86 out of 100AI agent readability56 out of 100 Top issueProduct structured data only appears after JavaScript runs 5 products rebuilt with fixes 437 productsRead the audit →
allbirds.com
Allbirds
Ad feed health75 out of 100AI agent readability44 out of 100 Top issueNo material, measurements or specs in the product data 5 products rebuilt with fixes 691 productsRead the audit →
deathwishcoffee.com
Death Wish Coffee
Ad feed health75 out of 100AI agent readability55 out of 100 Top issueNo material, measurements or specs in the product data 5 products rebuilt with fixes 147 productsRead the audit →
rothys.com
Rothy's
Ad feed health94 out of 100AI agent readability93 out of 100 Top issueMissing description 5 products rebuilt with fixes 749 productsRead the audit →
taylorstitch.com
Taylor Stitch
Ad feed health83 out of 100AI agent readability76 out of 100 Top issueNo material, measurements or specs in the product data 5 products rebuilt with fixes 1,974+ productsRead the audit → Your storeRun the same audit on your own catalog — free, read-only, in about a minute.Run free audit →Across the library
The issues we find most often
Counted across the 5 published audits. Each one links to what it costs and how to fix it.
- 5/5No material, measurements or specs in the product dataAgents filter on facts ("merino, under 300g, machine-washable"). With none, the product can't match those queries.
- 4/5Specs buried in prose, not structuredFacts exist in paragraphs but not as attributes; agents and feeds have to guess, and usually drop them.
- 3/5Missing descriptionDescription is a required feed attribute; agents have nothing to reason over.
- 3/5Very thin description (< 150 characters)Short copy limits which searches the item can match in Shopping and gives agents nothing to quote.
- 3/5No product typeWithout a product type, mapping to Google's taxonomy (and bidding by category) falls back to guesswork.
- 3/5Apparel item: color / size / gender / age group not publishedGoogle requires these for apparel in key markets (US, UK, DE, FR, JP, BR), and agents filter on them. If they only live in app metafields, no crawler or agent can see them.
Reading the scores
Two scores, out of 100
Ad feed health measures what gets products disapproved or throttled in Google Shopping and Meta catalogs: invalid GTINs, prices and stock in structured data that don't match the store, missing apparel attributes, undersized images.
AI agent readability measures whether ChatGPT, Perplexity and Google AI Mode can read a product's facts — structured data that's visible without JavaScript, described variants, materials and specs as attributes, and crawlers that aren't blocked.
- 80–100Good: few products are at risk, and the remaining fixes are small.
- 50–79Needs work: a meaningful share of the catalog is disapproved, throttled or unreadable.
- 0–49Poor: most sampled products are missing facts that Google and AI agents need.
Get the same report for your store
Free, read-only, about a minute — with your five worst products rebuilt from your own photos and copy.
Run free audit