The Commerce Eligibility Index CEI) — How We Evaluate Your Catalog
A Score That Means Something
When we evaluate a product catalog, we produce a single number: The Commerce Eligibility Index (CEI), scored from 0 to 100. It measures how well your catalog is structured for AI-mediated commerce — not just whether it looks good to humans, but whether machines can read it, trust it, and recommend from it with confidence.
But unlike a vanity metric, the CEI is built from the ground up with evidence. Every point in your score traces back to specific products, specific attributes, and specific structural patterns we can show you.
Here's what goes into it.
Five Pillars, One Score
We evaluate your catalog across five dimensions. Each one measures a different aspect of AI readiness:
Foundation
Can AI systems parse your catalog at all? We look at whether your product listings have complete, well-structured attributes — the basic building blocks that AI agents need to understand what you sell. Missing fields, sparse descriptions, and inconsistent formatting all show up here.
We also assess your visual content. Product images need to be clear, properly described, and varied enough to give AI systems visual context — not just a single hero shot repeated across colorways.
Differentiation
When AI compares your products to each other — and to competitors — can it tell them apart? We measure how semantically distinct your product descriptions are, whether your catalog surfaces meaningful tradeoffs between similar items, and whether your products cluster into coherent, well-separated groups.
If every product in your catalog uses the same generic description template, AI systems can't distinguish between them. Differentiation measures whether your catalog data helps or hinders product-level recommendations.
Retrieval
This is the acid test: when someone asks a natural-language shopping question, does the right product come back?
We simulate the kind of queries real shoppers ask — "waterproof hiking boots under $200," "organic cotton baby clothes," "wireless earbuds with noise cancellation" — and measure how well your catalog responds. We also test whether your products satisfy the specific constraints in those queries. If someone asks for "under $200" and your catalog returns a $250 product, that's a retrieval failure.
Integrity
Can AI trust what your catalog says? We look for factual accuracy, checking whether product claims are supported by the underlying data. If a description says "made from 100% recycled materials" but the material field says "polyester," that's a discrepancy an AI system will notice — and penalize.
We also analyze what your own customers are saying. Review sentiment on your owned platforms tells AI systems whether the products live up to their descriptions.
Authority
How does the broader ecosystem see your brand? We measure your visibility in independent reviews, your presence in brand discussions across the web, and — critically — whether AI shopping assistants are already recommending you.
That last one matters. We can actually test whether systems like ChatGPT, Claude, and Gemini include your brand in their shopping recommendations. If they don't, we can often tell you why.
How We Score It
Each pillar contributes a weighted share to your overall CEI score. The weights reflect how important each dimension is for AI commerce readiness — Foundation and Differentiation carry the most weight because they're the prerequisites. Without clean, structured, differentiated data, the other pillars don't matter.
One important design choice: we use a scoring method that penalizes zero-pillar scores heavily. You can't compensate for terrible Foundation quality with excellent Authority. If any pillar is near zero, your overall score drops significantly. This is intentional — it reflects reality. An AI agent won't recommend products it can't parse, no matter how many positive reviews exist.
What You Get
Every CatalogSignal engagement produces a diagnostic package designed for both technical teams and executive leadership:
Your CEI Score — The overall number (0-100) plus sub-scores for each of the five pillars. You can see exactly where you're strong and where you're exposed.
Top Risks and Quick Wins — We identify the five biggest structural risks in your catalog and the five changes that will have the most impact. Each one is backed by specific product examples — not abstract advice, but "here are the 47 SKUs with this problem."
Moveability Matrix — A visual map showing which improvements give you the biggest score lift for the least effort. Some fixes are high-impact and easy (fix missing material fields across 200 products). Others are high-impact but hard (rewrite 3,000 product descriptions). We help you prioritize.
Revenue Exposure Model — We model the financial impact of catalog gaps across three AI adoption scenarios (conservative, moderate, aggressive). This translates "your catalog has structural problems" into "here's what this could cost you" — language that resonates in the boardroom.
Remediation Roadmap — A three-tier action plan: Fix First (days), Next 30 Backlog (weeks), and Strategic Plays (months). Each recommendation is specific, actionable, and prioritized.
What Makes This Different
We test the data, not the search engine. Traditional SEO audits measure how well search engines index your pages. We measure whether your underlying product data is structured well enough for AI systems to reason about. These are fundamentally different problems.
We measure, we don't survey. Your CEI score is computed from your actual catalog data using consistent, repeatable methods. No subjectivity, no interpretation bias.
We show our work. Every metric comes with evidence. When we say your Retrieval pillar scored 62, we can show you exactly which queries failed, which products were incorrectly returned, and what structural patterns caused the mismatches.
We tune for your vertical. A fashion brand's catalog has different structural requirements than an electronics retailer's. Our evaluation adapts to your industry — the signals we measure are the same, but the benchmarks and expectations are calibrated for your competitive context.
We track progress over time. The CEI isn't a one-time audit. Run it quarterly or after major catalog updates to measure whether your investments in data quality are actually moving the needle. Compare yourself against industry benchmarks to see where you stand relative to competitors.
The Bottom Line
Your product catalog was built for humans. The next generation of commerce will be mediated by AI. The gap between those two realities is measurable — and closeable.
The Commerce Eligibility Index gives you the measurement, the diagnosis, and the roadmap. What you do with it is up to you.
The Commerce Eligibility Index is an AI readiness diagnostic for retail product catalogs, built by Catalog Signals and powered Bubo AI. Request an assessment at catalogsignal.com.