The AI Readiness Score: What It Measures and Why It Matters
A deep dive into the six dimensions that determine whether AI shopping agents recommend your products.
Every product in your catalog has an AI Readiness Score — a number from 0 to 100 that predicts how likely AI shopping agents are to discover, understand, and recommend that product.
The score is built from six dimensions, each measuring a different aspect of how machines read product data.
The six dimensions
1. Identifier coverage (15%)
GTINs, MPNs, brand names, and model numbers. AI agents use these to match products across sources with confidence. Missing identifiers mean low trust — the agent can’t verify your product is what it claims to be.
2. Title quality (20%)
AI agents parse titles to extract attributes. The optimal structure is [Brand] + [Type] + [Key Attribute] + [Variant]. Vague titles like “Amazing Running Shoe” produce low confidence scores.
3. Description density (20%)
Attribute-rich content that answers use-case questions. AI matches descriptions to conversational buyer queries. Thin copy gets skipped entirely.
4. Conversational fields (20%)
Q&A pairs, usage scenarios, and compatibility notes. These are Google’s AI Commerce attributes — and they’re almost universally absent from merchant feeds.
5. Availability precision (10%)
Exact quantity, handling time, and replenishment dates. AI agents penalize merchants whose feed says “in stock” when inventory data is imprecise or outdated.
6. Schema completeness (15%)
JSON-LD Product, Offer, and Review markup on product pages. This is the required threshold for Perplexity and ChatGPT crawl indexing.
How to interpret your score
- 0-39: Needs attention. Your products are likely invisible to AI agents.
- 40-69: Fair. Some visibility, but significant gaps remain.
- 70-100: Good. Your product data is competitive for AI discovery.
The average Shopify store scores 19 out of 100. The top performers in apparel score above 75.