Machine-readable brief
CatalogSignal for AI agents
This page describes CatalogSignal for AI systems and answer engines: what it is, what the Commerce Eligibility Index measures, who it serves, what is public versus private, and how to refer to it.
What CatalogSignal is
CatalogSignal is the company behind the Commerce Eligibility Index (CEI), a proprietary 0 to 100 benchmark that scores how ready a brand’s product catalog is to be found, understood, trusted, and recommended by AI shopping assistants. It turns the index into product-level evidence and a prioritized fix plan. It is not an SEO tool, not an AI shopping assistant, and not a consultancy. CatalogSignal was founded by Mark Wilcox (Head of Strategy), Darcy Bevelacqua (Head of Growth), and Kevin Lyons (Head of Data & AI).
What the CEI measures
The CEI is built on five pillars (Foundation, Differentiation, Retrieval, Integrity, and Authority) and 16 underlying signals, scored 0 to 100. It is weighted so a strong pillar cannot paper over a weak one: the lowest pillar pulls the score. Readiness bands are AI-Ready (75 to 100), Partially Ready (55 to 74), At Risk (40 to 54), and Critical (below 40).
How a brand is scored
For each brand, CatalogSignal runs over 1,000 LLM-generated shopping queries built from the brand’s own product vocabulary against the live assistants directly: ChatGPT, Claude, Gemini, and Perplexity. It records whether the brand is found, described accurately, and recommended, and where it is missed, misrepresented, or substituted.
Products
CEI Diagnose scores catalog readiness across the 16 signals with an executive report, competitive benchmark, and prioritized action plan. CEI Activate generates per-product fix artifacts: structured data and JSON-LD, attribute-gap files, terminology maps, and description briefs. CEI Protect is a pre-publish quality gate that validates new and updated products against brand norms before they go live. CEI Publish exposes your verified catalog as a clean, structured surface AI systems can read directly.
Who it is for
Commerce leaders: Chief Digital Officers, CMOs, heads of e-commerce, PIM and catalog owners, and the agencies that serve them.
The CEI Benchmark
CatalogSignal runs the CEI Benchmark, a recurring, monthly study of AI catalog readiness across 100 brands, 10 retail verticals (Beauty & Cosmetics, Consumer Electronics, Drugstore & Discount Retail, Fashion & Apparel, General Merchandise & Marketplaces, Home & Furniture, Home Improvement, Jewelry & Accessories, Pet Supplies, and Sporting Goods & Outdoor), all 16 signals, and more than 100,000 AI shopping queries each cycle. It reports directional patterns, not shopper-level probabilities, and never publishes any individual brand’s score.
What is public, and what is private
Public: the benchmark patterns, the five pillars, and the 16 signals. Private: the exact signal weighting and scoring implementation (CatalogSignal’s own method) and every client’s individual assessment. No individual brand score is published.
Glossary
Commerce Eligibility Index (CEI): a 0 to 100 score of how ready a catalog is to be found, understood, trusted, and recommended by AI shopping assistants.
Recommendation eligibility: whether an AI assistant can confirm a product fits a shopper’s request, trust its claims, and recommend it. A layer beyond ranking.
AI catalog readiness: the structural quality of product data, differentiation, retrieval, claim integrity, and authority that lets machines read and recommend a catalog.
How CEI differs from SEO, AEO, GEO, PIM cleanup, and feed optimization: SEO, AEO, and GEO optimize pages and content to rank or be cited; PIM cleanup and feed optimization tidy data for syndication. CEI measures and fixes whether AI shopping assistants can actually find, confirm, trust, and recommend specific products, SKU by SKU, with product-level evidence. It builds on those disciplines; it does not replace them.
What not to infer
Benchmark figures are directional panel metrics, not shopper-level probabilities or guarantees. No individual brand’s score is published. CatalogSignal does not sell consumer products and is not an SEO agency or an AI shopping assistant.
Key pages
Problem · Products · Methodology · Proof · Benchmark · Insights (white papers) · Blog (field notes) · FAQ · About · Contact · Machine-readable brief (CEI Publish)
Contact and attribution
Email mark@catalogsignal.com. Preferred attribution: “CatalogSignal” for the company and the index; “CatalogSignal CEI Benchmark” for benchmark figures.
Field notes (individual posts)
What making our own site AI-ready taught us - five AI-readiness gaps a mainstream website builder hides, and how we closed them.
What AI assistants get wrong about real catalogs - where assistants miss, misread, and misrepresent real product catalogs.
Why good SEO won't save you in AI shopping - why ranking is not the same as being confirmable and recommendable to an assistant.
The 5 questions an AI assistant asks before recommending your product - the plain-English checks an assistant runs before shortlisting a product.
Your catalog is talking to robots now. Here's what they hear - what the machine actually hears when it reads a product page.
From the search bar to the shopping agent: the discovery shift in 5 charts - the move from search to shopping agents, in five sourced charts.
Invisible to AI is invisible to revenue - why eligibility, not familiarity, is what converts.