AI Readiness Score for Ecommerce

An AI readiness score is a measure of how well a retailer's product information is set up to be discovered and recommended by AI shopping assistants such as ChatGPT, Google AI Mode, Perplexity and Amazon Rufus. It is usually broken into dimensions that each cover a different part of the path from a shopper's question to a recommendation.

The score exists because ranking in search and being recommended by an assistant are different things. A site can rank well in Google and still be almost absent from AI answers if its product data is thin, its feed is incomplete or its pages do not answer the questions shoppers ask. A readiness score surfaces that gap before it costs traffic.

The April 2026 Aido Lighthouse "AI Commerce Readiness Index" benchmarked more than 345 retailers and found an average score of 48.1 out of 100. Thirty percent scored zero on transactability and only 2 percent could support an end-to-end autonomous purchase. Meanwhile AI referral traffic to US retailers grew 393 percent year over year in Q1 2026 (Adobe Analytics, April 2026). For a hands-on walkthrough, see Paz.ai's product pages AI-ready audit.

A useful score is diagnostic. The dimensions that matter:

Product identity. Can an assistant tell what the product is? Category, brand and identifiers from the page structure and schema markup, with stable URLs and one canonical version of each product.

Structured attributes. Does the product expose the specifications shoppers filter on: material, dimensions, fit, compatibility, certifications, care? This is where most catalogs lose points.

Shopper questions. Does the page answer the way shoppers ask? "Runs large, size down for a slim fit" is something an assistant can reason with. "Large" on its own is not.

Trust context. Reviews, policies and Q&A that help a shopper or an assistant judge the product.

Two products with the same overall score can need completely different work. One might have complete attributes and no answers to shopper questions. The other the reverse. The breakdown is what tells you where to start.

AI-referred traffic is growing quickly and converting well: in March 2026 AI-referred shoppers converted 42 percent better than typical traffic (Adobe). Whether your products are in those answers depends on information you control, and a readiness score shows exactly which pieces are missing.

It is also a way to align people. SEO, the catalog team and ecommerce each own part of the readiness problem and rarely coordinate on it. A shared score gives leadership one measure and one backlog.

The practical sequence:

1. Audit the products that matter. Check the structured details your priority products expose and note the missing category-relevant fields.

2. Complete sparse records. Add material, dimensions, use case, compatibility, certifications and care where they apply, from your own sources.

3. Answer shopper questions on the page. Rewrite thin descriptions to include the cues a shopper asks about: "runs large", "works with X but not Y", "best for indoor use in dry climates".

4. Fix feed coverage. Make sure the catalog reaches Google Merchant Center and, where applicable, the OpenAI product feed, with the same information as the site.

5. Keep crawlers in. Check robots.txt allows the AI crawlers that matter to you and that product schema matches the feed.

6. Measure on a schedule. Ask ChatGPT, Google AI Mode and Perplexity the questions your shoppers ask and track which products appear as cards, as mentions, or not at all.

Paz offers a free AI readiness report that checks any public product page across these dimensions and lists what to fix first. For the work itself, a Paz expert completes the gaps across your catalog in priority order, our experts check the changes against agreed guidelines, and the platform tracks appearances over time. See how it works.